diff --git a/Streamlit/base_app.py b/Streamlit/base_app.py
index c27e4b4c..d1dfd1d9 100644
--- a/Streamlit/base_app.py
+++ b/Streamlit/base_app.py
@@ -29,11 +29,17 @@
 import pandas as pd
 
 # Vectorizer
-#news_vectorizer = open("streamlit/tfidfvect.pkl","rb")
-#test_cv = joblib.load(news_vectorizer) # loading your vectorizer from the pkl file
+news_vectorizer = open("vectorizer.pkl","rb")
+test_cv = joblib.load(news_vectorizer) # loading your vectorizer from the pkl file
 
 # Load your raw data
-#raw = pd.read_csv("streamlit/train.csv")
+raw = pd.read_csv("https://raw.githubusercontent.com/Jana-Liebenberg/2401PTDS_Classification_Project/main/Data/processed/train.csv", encoding="utf-8")
+
+# Load your machine learning models
+# Example: Replace 'model1.pkl' and 'model2.pkl' with your actual model file paths
+models = {
+	"Model 1": joblib.load("best_model.pkl")
+}
 
 # The main function where we will build the actual app
 def main():
@@ -53,7 +59,7 @@ def main():
 	if selection == "Information":
 		st.info("General Information")
 		# You can read a markdown file from supporting resources folder
-		st.markdown("Some information here")
+		st.markdown("This app uses machine learning models to classify news articles into predefined categories.")
 
 		
 	# Building out the predication page
@@ -62,18 +68,22 @@ def main():
 		# Creating a text box for user input
 		news_text = st.text_area("Enter Text","Type Here")
 
-		if st.button("Classify"):
+	if st.button("Classify"):
+		if news_text:
 			# Transforming user input with vectorizer
 			vect_text = test_cv.transform([news_text]).toarray()
-			# Load your .pkl file with the model of your choice + make predictions
-			# Try loading in multiple models to give the user a choice
-			predictor = joblib.load(open(os.path.join("streamlit/Logistic_regression.pkl"),"rb"))
-			prediction = predictor.predict(vect_text)
-
-			# When model has successfully run, will print prediction
-			# You can use a dictionary or similar structure to make this output
-			# more human interpretable.
-			st.success("Text Categorized as: {}".format(prediction))
+
+			# Get predictions from each model
+			predictions = {model_name: model.predict(vect_text)[0] for model_name, model in models.items()}
+
+			# Display predictions
+			st.subheader("Model Predictions:")
+			for model_name, category in predictions.items():
+				st.write(f"**{model_name}:** {category}")
+			# Show success message with categorized text
+			st.success("Text Categorized as: {}".format(", ".join(predictions.values())))
+		else:
+			st.warning("Please enter an article for classification.")
 
 # Required to let Streamlit instantiate our web app.  
 if __name__ == '__main__':
diff --git a/Streamlit/best_model.pkl b/Streamlit/best_model.pkl
new file mode 100644
index 00000000..764dfa43
Binary files /dev/null and b/Streamlit/best_model.pkl differ
diff --git a/Streamlit/kelly.ipynb b/Streamlit/kelly.ipynb
new file mode 100644
index 00000000..b564b509
--- /dev/null
+++ b/Streamlit/kelly.ipynb
@@ -0,0 +1,1257 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "40d42af3-0d36-4fed-ade8-5c189469bd8b",
+   "metadata": {},
+   "source": [
+    "<h1 style=\"color:black;\">Classification Project: Newspaper Articles</h1>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a4d699a7-050f-498c-a58f-8b8917905cf1",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">PROJECT OVERVIEW</h3>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0c22ed09-70b7-47bd-b81d-483344afa3bb",
+   "metadata": {},
+   "source": [
+    ">The key goal of this classification project is for our team to develop machine learning models that can automatically classify newspaper articles. This project is an end-to-end project covering all stages, from loading and preprocessing data to training models, evaluating them, and deploying the solution as a web app using Streamlit."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a830a160-194a-47a3-b888-e76eda1ee776",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">IMPORTING PACKAGES</h3>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "72bc717e-0601-441a-adcd-a96530c25810",
+   "metadata": {},
+   "source": [
+    "> This project will utilise the packages below to efficiently load, clean, analyse and visualise the newspaper article data. The packages are imported and necessary libraries are downloaded below:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "e73e73b6-d9d1-45a8-9aed-f1908111f145",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Importing Pandas for data manipulation and analysis\n",
+    "import pandas as pd\n",
+    "\n",
+    "# Importing NumPy for numerical operations\n",
+    "import numpy as np\n",
+    "\n",
+    "# Importing Matplotlib for data visualization\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Importing Seaborn for advanced data visualization\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# Importing IPython.display for displaying rich media in Jupyter Notebooks\n",
+    "from IPython.display import display, Image\n",
+    "\n",
+    "# Importing nltk for tokenization, lemmatization, and text preprocessing\n",
+    "import nltk \n",
+    "\n",
+    "# Importing re for regular expressions to clean and manipulate text\n",
+    "import re\n",
+    "\n",
+    "# Importing string for handling common string operations, such as removing punctuation\n",
+    "import string\n",
+    "\n",
+    "# Importing nltk.corpus to improve text processing\n",
+    "from nltk.corpus import  stopwords\n",
+    "\n",
+    "# Importing nltk.tokenize to split text into words or sentences for NLP tasks\n",
+    "from nltk.tokenize import word_tokenize, TreebankWordTokenizer\n",
+    "\n",
+    "# Importing nltk.stem to converts words to their root forms, improving text understanding\n",
+    "from nltk.stem import WordNetLemmatizer\n",
+    "\n",
+    "# Convert text to numerical vectors\n",
+    "from sklearn.feature_extraction.text import TfidfVectorizer \n",
+    "\n",
+    "# Splitting data\n",
+    "from sklearn.model_selection import train_test_split \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "344af896-aabd-4703-b9f7-7bcf0793d4b5",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[nltk_data] Downloading package punkt to\n",
+      "[nltk_data]     C:\\Users\\kelly\\AppData\\Roaming\\nltk_data...\n",
+      "[nltk_data]   Package punkt is already up-to-date!\n",
+      "[nltk_data] Downloading package punkt_tab to\n",
+      "[nltk_data]     C:\\Users\\kelly\\AppData\\Roaming\\nltk_data...\n",
+      "[nltk_data]   Package punkt_tab is already up-to-date!\n",
+      "[nltk_data] Downloading package stopwords to\n",
+      "[nltk_data]     C:\\Users\\kelly\\AppData\\Roaming\\nltk_data...\n",
+      "[nltk_data]   Package stopwords is already up-to-date!\n",
+      "[nltk_data] Downloading package wordnet to\n",
+      "[nltk_data]     C:\\Users\\kelly\\AppData\\Roaming\\nltk_data...\n",
+      "[nltk_data]   Package wordnet is already up-to-date!\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "True"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "nltk.download(['punkt','punkt_tab','stopwords','wordnet'])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fd9ce86b-af85-49da-847b-af99ed6f0591",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">DATA LOADING</h3>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c086f70a-7ec8-4d82-a919-9e8b5dd101e1",
+   "metadata": {},
+   "source": [
+    "> When the required packages are imported, for loading a CSV file, this Project will utilise Pandas to load the data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "23b26a61-f885-4289-9b76-719e36939eb3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load training and testing datasets\n",
+    "train_df = pd.read_csv(\"https://raw.githubusercontent.com/Jana-Liebenberg/2401PTDS_Classification_Project/main/Data/processed/train.csv\", encoding=\"utf-8\")\n",
+    "test_df = pd.read_csv(\"https://raw.githubusercontent.com/Jana-Liebenberg/2401PTDS_Classification_Project/main/Data/processed/test.csv\", encoding=\"utf-8\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "dd1b96ba-8e6f-4b36-829d-682ca5591500",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>headlines</th>\n",
+       "      <th>description</th>\n",
+       "      <th>content</th>\n",
+       "      <th>url</th>\n",
+       "      <th>category</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>RBI revises definition of politically-exposed ...</td>\n",
+       "      <td>The central bank has also asked chairpersons a...</td>\n",
+       "      <td>The Reserve Bank of India (RBI) has changed th...</td>\n",
+       "      <td>https://indianexpress.com/article/business/ban...</td>\n",
+       "      <td>business</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>NDTV Q2 net profit falls 57.4% to Rs 5.55 cror...</td>\n",
+       "      <td>NDTV's consolidated revenue from operations wa...</td>\n",
+       "      <td>Broadcaster New Delhi Television Ltd on Monday...</td>\n",
+       "      <td>https://indianexpress.com/article/business/com...</td>\n",
+       "      <td>business</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           headlines  \\\n",
+       "0  RBI revises definition of politically-exposed ...   \n",
+       "1  NDTV Q2 net profit falls 57.4% to Rs 5.55 cror...   \n",
+       "\n",
+       "                                         description  \\\n",
+       "0  The central bank has also asked chairpersons a...   \n",
+       "1  NDTV's consolidated revenue from operations wa...   \n",
+       "\n",
+       "                                             content  \\\n",
+       "0  The Reserve Bank of India (RBI) has changed th...   \n",
+       "1  Broadcaster New Delhi Television Ltd on Monday...   \n",
+       "\n",
+       "                                                 url  category  \n",
+       "0  https://indianexpress.com/article/business/ban...  business  \n",
+       "1  https://indianexpress.com/article/business/com...  business  "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>headlines</th>\n",
+       "      <th>description</th>\n",
+       "      <th>content</th>\n",
+       "      <th>url</th>\n",
+       "      <th>category</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>NLC India wins contract for power supply to Ra...</td>\n",
+       "      <td>State-owned firm NLC India Ltd (NLCIL) on Mond...</td>\n",
+       "      <td>State-owned firm NLC India Ltd (NLCIL) on Mond...</td>\n",
+       "      <td>https://indianexpress.com/article/business/com...</td>\n",
+       "      <td>business</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>SBI Clerk prelims exams dates announced; admit...</td>\n",
+       "      <td>SBI Clerk Prelims Exam: The SBI Clerk prelims ...</td>\n",
+       "      <td>SBI Clerk Prelims Exam: The State Bank of Indi...</td>\n",
+       "      <td>https://indianexpress.com/article/education/sb...</td>\n",
+       "      <td>education</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           headlines  \\\n",
+       "0  NLC India wins contract for power supply to Ra...   \n",
+       "1  SBI Clerk prelims exams dates announced; admit...   \n",
+       "\n",
+       "                                         description  \\\n",
+       "0  State-owned firm NLC India Ltd (NLCIL) on Mond...   \n",
+       "1  SBI Clerk Prelims Exam: The SBI Clerk prelims ...   \n",
+       "\n",
+       "                                             content  \\\n",
+       "0  State-owned firm NLC India Ltd (NLCIL) on Mond...   \n",
+       "1  SBI Clerk Prelims Exam: The State Bank of Indi...   \n",
+       "\n",
+       "                                                 url   category  \n",
+       "0  https://indianexpress.com/article/business/com...   business  \n",
+       "1  https://indianexpress.com/article/education/sb...  education  "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Display first few rows of the Dataframes as tables\n",
+    "display(train_df.head(2))\n",
+    "display(test_df.head(2))\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "39df45e6-3e02-4ffe-9ad8-c52f018c2b99",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">DATA CLEANING AND TEXT PREPROCESSING</h3>\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "599b9afc-199a-45c6-8f7d-45ebb2c44a74",
+   "metadata": {},
+   "source": [
+    "> Text cleaning is essential for improving model accuracy. We'll look into standardizing text by converting everything to lowercase, removing punctuation and special characters, splitting text into individual words for further processing and removing common words that don't add much meaning."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "773bd76a-4efd-42e8-8481-716a2165ed82",
+   "metadata": {},
+   "source": [
+    "Let's start by printing a list of all categories that are present in this dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "5c2cdca8-7aee-4155-bb44-2a785573ab2e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['business', 'sports', 'entertainment', 'education', 'technology']\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Print list of unique newspaper categories\n",
+    "category_labels = list(train_df.category.unique())\n",
+    "print(category_labels)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "f0597312-d23f-499c-8e28-93b58cee11a4",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Define a color palette\n",
+    "colors = sns.color_palette('husl', len(train_df['category'].value_counts()))\n",
+    "\n",
+    "# Visualize the distribution of the categories with colored bars\n",
+    "train_df['category'].value_counts().plot(kind='bar', color=colors)\n",
+    "plt.xlabel('Category')\n",
+    "plt.ylabel('Count')\n",
+    "plt.title('Distribution of Category Labels')\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "d8a9b578-b9ff-41e6-a8cb-176a3815b0be",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(2000, 5)"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Shape of the dataset\n",
+    "train_df.shape\n",
+    "test_df.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "8d8b7a2e-1512-472b-886e-40e36e2d71d5",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Missing values in dataset:\n",
+      " headlines      0\n",
+      "description    0\n",
+      "content        0\n",
+      "url            0\n",
+      "category       0\n",
+      "dtype: int64\n",
+      "Missing values in dataset:\n",
+      " headlines      0\n",
+      "description    0\n",
+      "content        0\n",
+      "url            0\n",
+      "category       0\n",
+      "dtype: int64\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Check for missing values\n",
+    "print(\"Missing values in dataset:\\n\", train_df.isnull().sum())\n",
+    "print(\"Missing values in dataset:\\n\", test_df.isnull().sum())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "4a4df036-ea5e-4e65-b911-7db0ca3837f9",
+   "metadata": {},
+   "source": [
+    "There was no need to handle missing values as there were none"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "a568052e-f838-4074-bab2-076aa2577af0",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "No duplicates found in the dataset.\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Check and remove duplicates\n",
+    "train_df.drop_duplicates(inplace=True) \n",
+    "test_df.drop_duplicates(inplace=True)\n",
+    "\n",
+    "if train_df.duplicated().any():\n",
+    "    print(\"Duplicates are present in the dataset.\")\n",
+    "else:\n",
+    "    print(\"No duplicates found in the dataset.\")\n",
+    "\n",
+    "if test_df.duplicated().any():\n",
+    "    print(\"Duplicates are present in the dataset.\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "f4712212-47e3-4d3e-974c-121060f5a69c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define a function to preprocess text data\n",
+    "def format_data (text):\n",
+    "    \"\"\"\n",
+    "    Function to clean text:\n",
+    "    - Convert to lowercase\n",
+    "    - Replace URLs with a placeholder\n",
+    "    - Remove HTML tags\n",
+    "    - Remove punctuation and numbers\n",
+    "    - Tokenize and remove stopwords\n",
+    "    - Lemmatize words\n",
+    "    \"\"\"\n",
+    "    if isinstance(text, str):  \n",
+    "        text = text.lower()  # Convert to lowercase\n",
+    "        \n",
+    "        # Replace URLs with placeholder\n",
+    "        text = re.sub(r'http\\S+|www\\S+|https\\S+', 'url-web', text, flags=re.MULTILINE)\n",
+    "        \n",
+    "        # Remove HTML tags\n",
+    "        text = re.sub(r'<.*?>', '', text)\n",
+    "        \n",
+    "        # Remove punctuation and numbers\n",
+    "        text = text.translate(str.maketrans('', '', string.punctuation + '0123456789'))\n",
+    "        \n",
+    "        # Tokenize the text\n",
+    "        tokens = word_tokenize(text)\n",
+    "        \n",
+    "        # Remove stopwords\n",
+    "        tokens = [word for word in tokens if word not in stopwords.words('english')]\n",
+    "        \n",
+    "        # Lemmatize words\n",
+    "        lemmatizer = WordNetLemmatizer()\n",
+    "        tokens = [lemmatizer.lemmatize(word) for word in tokens]\n",
+    "        \n",
+    "        return ' '.join(tokens)\n",
+    "    \n",
+    "    return \"\"  # Handle missing or invalid data\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "3238d931-6132-424c-9c66-3d81c2df0298",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "                                           headlines  \\\n",
+      "0  rbi revise definition politicallyexposed perso...   \n",
+      "1  ndtv q net profit fall r crore impacted lower ...   \n",
+      "2  akasa air ‘ well capitalised ’ grow much faste...   \n",
+      "3  india ’ current account deficit decline sharpl...   \n",
+      "4       state borrowing cost soar highest far fiscal   \n",
+      "\n",
+      "                                         description  \\\n",
+      "0  central bank also asked chairperson chief exec...   \n",
+      "1  ndtvs consolidated revenue operation r crore r...   \n",
+      "2  initial share sale open public subscription ju...   \n",
+      "3  current account deficit cad per cent gdp usd b...   \n",
+      "4  price shot reflecting overall higher risk aver...   \n",
+      "\n",
+      "                                             content  \n",
+      "0  reserve bank india rbi changed definition poli...  \n",
+      "1  broadcaster new delhi television ltd monday re...  \n",
+      "2  homegrown server maker netweb technology india...  \n",
+      "3  india ’ current account deficit declined sharp...  \n",
+      "4  state forced pay nose weekly auction debt tues...  \n"
+     ]
+    }
+   ],
+   "source": [
+    "# Apply the function to each column in the training dataset\n",
+    "train_df[\"headlines\"] = train_df[\"headlines\"].apply(format_data)\n",
+    "train_df[\"description\"] = train_df[\"description\"].apply(format_data)\n",
+    "train_df[\"content\"] = train_df[\"content\"].apply(format_data)\n",
+    "\n",
+    "# Preview the preprocessed data\n",
+    "print(train_df[[\"headlines\", \"description\", \"content\"]].head())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "d4cfdc97",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "                                           headlines  \\\n",
+      "0  nlc india win contract power supply rajasthan ...   \n",
+      "1  sbi clerk prelim exam date announced admit car...   \n",
+      "2  golden globe michelle yeoh ferrell angela bass...   \n",
+      "3                oneplus nord r part new price cut ’   \n",
+      "4  adani family ’ partner used ‘ opaque ’ fund in...   \n",
+      "\n",
+      "                                         description  \\\n",
+      "0  stateowned firm nlc india ltd nlcil monday sai...   \n",
+      "1  sbi clerk prelim exam sbi clerk prelim exam te...   \n",
+      "2  barbie top nominee year followed closely oppen...   \n",
+      "3       new deal make oneplus nord easy purchase r k   \n",
+      "4  citing review file multiple tax haven internal...   \n",
+      "\n",
+      "                                             content  \n",
+      "0  stateowned firm nlc india ltd nlcil monday sai...  \n",
+      "1  sbi clerk prelim exam state bank indian sbi an...  \n",
+      "2  michelle yeoh ferrell angela bassett amanda se...  \n",
+      "3  review oneplus nord g praised balanced combina...  \n",
+      "4  million dollar invested publicly traded stock ...  \n"
+     ]
+    }
+   ],
+   "source": [
+    "# Apply the function to each column in the test dataset\n",
+    "test_df[\"headlines\"] = test_df[\"headlines\"].apply(format_data)\n",
+    "test_df[\"description\"] = test_df[\"description\"].apply(format_data)\n",
+    "test_df[\"content\"] = test_df[\"content\"].apply(format_data)\n",
+    "\n",
+    "# Preview the preprocessed test data\n",
+    "print(test_df[[\"headlines\", \"description\", \"content\"]].head())\n",
+    "# Combine the text columns into a single column for vectorization"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "506c1550",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">FEATURE ENGENEERING</h3>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "2ac4773c-a6be-49c8-836c-f772e7d5bef0",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "y_train shape: (5520,)\n"
+     ]
+    }
+   ],
+   "source": [
+    "import joblib\n",
+    "from sklearn.feature_extraction.text import CountVectorizer\n",
+    "\n",
+    "# Format target variables\n",
+    "y_train = train_df['category']\n",
+    "y_test = test_df['category']\n",
+    "\n",
+    "print('y_train shape:', y_train.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "c00d2ee9-f2b2-4a1a-b747-893a8964776c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Defining a function to vectorize data\n",
+    "def vectorize_data(column):\n",
+    "    # Initialize CountVectorizer with stop_words handling and minimum document frequency\n",
+    "    vect = CountVectorizer(stop_words='english', min_df=0.01)  \n",
+    "    X_train = vect.fit_transform(train_df[column]).toarray()  # Transform and convert to NumPy array\n",
+    "    X_test = vect.transform(test_df[column]).toarray()\n",
+    "    joblib.dump(vect, 'vectorizer.pkl')\n",
+    "\n",
+    "    return X_train, X_test"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "56d2b2ff-7f88-48e7-8ab8-f1a3f9785a9c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Apply vectorization to each column\n",
+    "X_headlines_train, X_headlines_test = vectorize_data('headlines')\n",
+    "X_description_train, X_description_test = vectorize_data('description')  \n",
+    "X_content_train, X_content_test = vectorize_data('content')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "79357e07",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "X_headlines_train: (5520, 108)\n",
+      "X_headlines_test: (2000, 108)\n",
+      "X_description_train: (5520, 192)\n",
+      "X_description_test: (2000, 192)\n",
+      "X_content_train: (5520, 1506)\n",
+      "X_content_test: (2000, 1506)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Print results\n",
+    "print('X_headlines_train:', X_headlines_train.shape)\n",
+    "print('X_headlines_test:', X_headlines_test.shape)\n",
+    "print('X_description_train:', X_description_train.shape)\n",
+    "print('X_description_test:', X_description_test.shape)\n",
+    "print('X_content_train:', X_content_train.shape)\n",
+    "print('X_content_test:', X_content_test.shape)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5da58dcc",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">TRAINING MODELS</h3>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "c846de84",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# !pip install mlflow when MLflow is not installed\n",
+    "# Set the tracking URI for MLflow\n",
+    "import mlflow\n",
+    "import mlflow.sklearn"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "56f64c18",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Experiment: artifact_location='file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026', creation_time=1746429792104, experiment_id='813981363149581026', last_update_time=1746429792104, lifecycle_stage='active', name='News Category Classification', tags={}>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mlflow.set_experiment(\"News Category Classification\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "ac32d330",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Import necessary libraries for model training\n",
+    "from sklearn.naive_bayes import MultinomialNB\n",
+    "from sklearn.metrics import classification_report, accuracy_score, confusion_matrix\n",
+    "from sklearn.pipeline import make_pipeline\n",
+    "from sklearn.linear_model import LogisticRegression \n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from sklearn.svm import SVC"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "64ffb887",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "X_train shape: (5520, 1806)\n",
+      "X_test shape: (2000, 1806)\n",
+      "X_train_split shape: (4416, 1806)\n",
+      "X_val_split shape: (1104, 1806)\n",
+      "y_train_split shape: (4416,)\n",
+      "y_val_split shape: (1104,)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Prepare the data for model training\n",
+    "#combine all three vectorized columns into a single feature set\n",
+    "X_train = np.concatenate((X_headlines_train, X_description_train, X_content_train), axis=1)\n",
+    "X_test = np.concatenate((X_headlines_test, X_description_test, X_content_test), axis=1)\n",
+    "\n",
+    "# Print the shape of the training and testing data\n",
+    "print('X_train shape:', X_train.shape)\n",
+    "print('X_test shape:', X_test.shape)\n",
+    "\n",
+    "# Split the training data into training and validation sets\n",
+    "X_train_split, X_val_split, y_train_split, y_val_split = train_test_split(X_train, y_train, test_size=0.2, random_state=42)\n",
+    "\n",
+    "# Print the shape of the training and validation sets   \n",
+    "print('X_train_split shape:', X_train_split.shape)\n",
+    "print('X_val_split shape:', X_val_split.shape)\n",
+    "print('y_train_split shape:', y_train_split.shape)\n",
+    "print('y_val_split shape:', y_val_split.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "1cb490b2",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Initialize the models\n",
+    "models = {\n",
+    "    'Logistic Regression': LogisticRegression(max_iter=1000),\n",
+    "    'Random Forest': RandomForestClassifier(n_estimators=100, random_state=42),\n",
+    "    'Support Vector Classifier': SVC(kernel='linear', random_state=42),\n",
+    "    'Naive Bayes': MultinomialNB()\n",
+    "}\n",
+    "# Initialize a dictionary to store the results\n",
+    "results = {}\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ef32b136",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">TRAINING AND EVALUATING MODELS INSIDE MLFLOW</h3>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "e2073997",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: Logistic Regression\n",
+      "Classification Report:\n",
+      "                precision    recall  f1-score   support\n",
+      "\n",
+      "     business       0.94      0.96      0.95       245\n",
+      "    education       0.99      0.99      0.99       274\n",
+      "entertainment       1.00      0.99      1.00       178\n",
+      "       sports       0.99      0.99      0.99       137\n",
+      "   technology       0.97      0.96      0.96       270\n",
+      "\n",
+      "     accuracy                           0.97      1104\n",
+      "    macro avg       0.98      0.98      0.98      1104\n",
+      " weighted avg       0.97      0.97      0.97      1104\n",
+      "\n",
+      "Confusion Matrix:\n",
+      " [[236   2   0   0   7]\n",
+      " [  3 270   0   0   1]\n",
+      " [  0   0 177   0   1]\n",
+      " [  2   0   0 135   0]\n",
+      " [ 10   1   0   1 258]]\n",
+      "Accuracy: 0.9746376811594203\n",
+      "--------------------------------------------------\n",
+      "Model: Random Forest\n",
+      "Classification Report:\n",
+      "                precision    recall  f1-score   support\n",
+      "\n",
+      "     business       0.96      0.96      0.96       245\n",
+      "    education       0.99      0.99      0.99       274\n",
+      "entertainment       0.99      0.95      0.97       178\n",
+      "       sports       0.96      0.96      0.96       137\n",
+      "   technology       0.93      0.95      0.94       270\n",
+      "\n",
+      "     accuracy                           0.96      1104\n",
+      "    macro avg       0.97      0.96      0.96      1104\n",
+      " weighted avg       0.96      0.96      0.96      1104\n",
+      "\n",
+      "Confusion Matrix:\n",
+      " [[235   1   0   0   9]\n",
+      " [  0 272   0   1   1]\n",
+      " [  0   1 169   1   7]\n",
+      " [  2   0   0 132   3]\n",
+      " [  7   2   2   3 256]]\n",
+      "Accuracy: 0.9637681159420289\n",
+      "--------------------------------------------------\n",
+      "Model: Support Vector Classifier\n",
+      "Classification Report:\n",
+      "                precision    recall  f1-score   support\n",
+      "\n",
+      "     business       0.95      0.97      0.96       245\n",
+      "    education       0.99      0.99      0.99       274\n",
+      "entertainment       1.00      0.98      0.99       178\n",
+      "       sports       0.98      0.99      0.98       137\n",
+      "   technology       0.96      0.96      0.96       270\n",
+      "\n",
+      "     accuracy                           0.97      1104\n",
+      "    macro avg       0.98      0.98      0.98      1104\n",
+      " weighted avg       0.97      0.97      0.97      1104\n",
+      "\n",
+      "Confusion Matrix:\n",
+      " [[237   1   0   0   7]\n",
+      " [  1 270   0   0   3]\n",
+      " [  0   0 175   2   1]\n",
+      " [  2   0   0 135   0]\n",
+      " [  9   1   0   1 259]]\n",
+      "Accuracy: 0.9746376811594203\n",
+      "--------------------------------------------------\n",
+      "Model: Naive Bayes\n",
+      "Classification Report:\n",
+      "                precision    recall  f1-score   support\n",
+      "\n",
+      "     business       0.96      0.94      0.95       245\n",
+      "    education       1.00      0.99      0.99       274\n",
+      "entertainment       0.98      0.99      0.98       178\n",
+      "       sports       0.97      0.97      0.97       137\n",
+      "   technology       0.94      0.96      0.95       270\n",
+      "\n",
+      "     accuracy                           0.97      1104\n",
+      "    macro avg       0.97      0.97      0.97      1104\n",
+      " weighted avg       0.97      0.97      0.97      1104\n",
+      "\n",
+      "Confusion Matrix:\n",
+      " [[230   0   1   0  14]\n",
+      " [  2 271   1   0   0]\n",
+      " [  0   0 176   1   1]\n",
+      " [  1   0   2 133   1]\n",
+      " [  6   1   0   3 260]]\n",
+      "Accuracy: 0.9692028985507246\n",
+      "--------------------------------------------------\n"
+     ]
+    }
+   ],
+   "source": [
+    "# End any active MLflow run\n",
+    "mlflow.end_run() \n",
+    "\n",
+    "# Train and evaluate each model inside MLflow tracking\n",
+    "results = {}\n",
+    "\n",
+    "for model_name, model in models.items():\n",
+    "    with mlflow.start_run(run_name=model_name, nested=True):  # Only use nested runs\n",
+    "        # Train the model\n",
+    "        model.fit(X_train_split, y_train_split)\n",
+    "        \n",
+    "        # Make predictions on the validation set\n",
+    "        y_val_pred = model.predict(X_val_split)\n",
+    "        \n",
+    "        # Calculate accuracy and store results\n",
+    "        accuracy = accuracy_score(y_val_split, y_val_pred)\n",
+    "        results[model_name] = accuracy\n",
+    "        \n",
+    "        # Log the model and metrics to MLflow\n",
+    "        mlflow.log_param(\"model_name\", model_name)\n",
+    "        mlflow.log_metric(\"accuracy\", accuracy)\n",
+    "        \n",
+    "        # Print classification report and confusion matrix\n",
+    "        print(f\"Model: {model_name}\")\n",
+    "        print(\"Classification Report:\\n\", classification_report(y_val_split, y_val_pred))\n",
+    "        print(\"Confusion Matrix:\\n\", confusion_matrix(y_val_split, y_val_pred))\n",
+    "        print(\"Accuracy:\", accuracy)\n",
+    "        print(\"-\" * 50)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0c79be29",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">SELECTING THE BEST MODEL</h3>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "32cacfb8",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Best Model: Logistic Regression\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2025/05/05 11:18:12 WARNING mlflow.models.model: Model logged without a signature and input example. Please set `input_example` parameter when logging the model to auto infer the model signature.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Best model performance on test set:\n",
+      "Accuracy: 0.975\n",
+      "Classification Report:\n",
+      "                precision    recall  f1-score   support\n",
+      "\n",
+      "     business       0.97      0.96      0.97       400\n",
+      "    education       0.99      0.99      0.99       400\n",
+      "entertainment       0.99      0.98      0.99       400\n",
+      "       sports       0.98      0.98      0.98       400\n",
+      "   technology       0.94      0.96      0.95       400\n",
+      "\n",
+      "     accuracy                           0.97      2000\n",
+      "    macro avg       0.98      0.97      0.98      2000\n",
+      " weighted avg       0.98      0.97      0.98      2000\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "#selecting the best model\n",
+    "best_model = max(results, key=results.get)\n",
+    "print(f\"Best Model: {best_model}\")\n",
+    "\n",
+    "# Train the best model on the entire training set\n",
+    "best_model_instance = models[best_model]\n",
+    "best_model_instance.fit(X_train, y_train)  \n",
+    "\n",
+    "# Save the best model to MLflow\n",
+    "with mlflow.start_run():\n",
+    "    mlflow.sklearn.log_model(best_model_instance, \"Best_model\")\n",
+    "\n",
+    "# Make predictions on the test set\n",
+    "y_test_pred = best_model_instance.predict(X_test)\n",
+    "\n",
+    "# Evaluate the model on the test set\n",
+    "accuracy_test = accuracy_score(y_test, y_test_pred)\n",
+    "mlflow.log_metric(\"test_accuracy\", accuracy_test)\n",
+    "report_test = classification_report(y_test, y_test_pred)    \n",
+    "\n",
+    "# Print the test set results\n",
+    "print(\"Best model performance on test set:\")\n",
+    "print(f\"Accuracy: {accuracy_test}\")\n",
+    "print(\"Classification Report:\\n\", report_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "b23a6331",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>category</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>business</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>education</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>entertainment</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>technology</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>business</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        category\n",
+       "0       business\n",
+       "1      education\n",
+       "2  entertainment\n",
+       "3     technology\n",
+       "4       business"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Save the model using joblib\n",
+    "joblib.dump(best_model_instance, 'best_model.pkl')\n",
+    "\n",
+    "# Save the predictions to a CSV file\n",
+    "predictions_df = pd.DataFrame({'category': y_test_pred})\n",
+    "predictions_df.to_csv('predictions.csv', index=False)\n",
+    "\n",
+    "# Display the first few rows of the predictions DataFrame\n",
+    "display(predictions_df.head())  "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "id": "0674fb87",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x800 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "#confusion matrix visualization\n",
+    "\n",
+    "# Create a confusion matrix\n",
+    "cm = confusion_matrix(y_test, y_test_pred, labels=category_labels)\n",
+    "\n",
+    "# Plot the confusion matrix\n",
+    "plt.figure(figsize=(10, 8))\n",
+    "sns.heatmap(cm, annot=True, fmt='d', cmap='coolwarm', xticklabels=category_labels, yticklabels=category_labels)\n",
+    "plt.xlabel('Predicted')\n",
+    "plt.ylabel('True')\n",
+    "plt.title('Confusion Matrix')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5def9ed5",
+   "metadata": {},
+   "source": [
+    "This confusion matrix visually represents the performance of a classification model in categorizing various topics: business, sports, entertainment, education, and technology. The diagonal values show correct classifications, with Education having the greater correctly classified number and only 4 samples being incorrectly classified( one classified as business and 3 classified as technology)\n",
+    "\n",
+    "**solutions to misclassifications:**\n",
+    "Our confusion matrix visualization suggest there is a confusion between certain categories, which may require better feature engineering or tuning of the model."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "id": "43a36521",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# loading saved model and vectorizer and making predictions on new data\n",
+    "loaded_model = joblib.load(\"best_model.pkl\")\n",
+    "loaded_vectorizer = joblib.load(\"vectorizer.pkl\")\n",
+    "\n",
+    "def predict_category(headlines, description, content):\n",
+    "    \"\"\"\n",
+    "    Function to predict the category of a news article.\n",
+    "    \"\"\"\n",
+    "    headlines = format_data(headlines)\n",
+    "    description = format_data(description)\n",
+    "    content = format_data(content)\n",
+    "\n",
+    "    # Vectorize input\n",
+    "    X_new = np.concatenate((\n",
+    "        loaded_vectorizer.transform([headlines]).toarray(),\n",
+    "        loaded_vectorizer.transform([description]).toarray(),\n",
+    "        loaded_vectorizer.transform([content]).toarray()\n",
+    "    ), axis=1)\n",
+    "\n",
+    "    return loaded_model.predict(X_new)[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "id": "4154bdea",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "^C\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Run MLflow UI to view the experiment results\n",
+    "!mlflow ui "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a05af281-606b-47a0-b916-3277cd1c7992",
+   "metadata": {},
+   "source": [
+    "<h3 style=\"color:blue;\">CONCLUSION AND INSIGHTS</h3>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ecb3bd5b-ab4f-4c3b-99d6-cd17fb23a6f0",
+   "metadata": {},
+   "source": [
+    "> To conclude, our team built a comprehensive news categorization pipeline, from preprocessing text to training multiple models and evaluating performance using MLflow. The key steps that were taken were: \n",
+    "\n",
+    "**1.** Data Preparation – Combined vectorized text features & split into training, validation, and test sets.\n",
+    "\n",
+    "**2.** Model Training & Evaluation – Trained Logistic Regression, Random Forest, Naïve Bayes, and SVM, logging accuracy & metrics in MLflow.\n",
+    "\n",
+    "**3.** Best Model Selection – Identified the best-performing model dynamically based on validation accuracy.\n",
+    "\n",
+    "**4.** Model Deployment – Saved the best model and vectorizer for future predictions using joblib.\n",
+    "\n",
+    "**5.** Results Tracking – MLflow logged parameters, accuracy, and confusion matrices, allowing comparison between models.\n",
+    "\n",
+    "**6.** Created a Prediction Function – Built a reusable function to classify new articles dynamically.\n",
+    "\n",
+    "**Results from MLflow:**\n",
+    "The \"Test Accuracy\" showed that the best-performing model wasthe Logistic Regression model\n",
+    ", which achieved strong accuracy, with performance tracked across all trained models.\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "base",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/Streamlit/mlruns/0/meta.yaml b/Streamlit/mlruns/0/meta.yaml
new file mode 100644
index 00000000..90c86f03
--- /dev/null
+++ b/Streamlit/mlruns/0/meta.yaml
@@ -0,0 +1,6 @@
+artifact_location: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/0
+creation_time: 1746429792088
+experiment_id: '0'
+last_update_time: 1746429792088
+lifecycle_stage: active
+name: Default
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/meta.yaml b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/meta.yaml
new file mode 100644
index 00000000..7ea802b2
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/artifacts
+end_time: 1746436637371
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 0d7abd2ef2804a57b1754adba66adb00
+run_name: Logistic Regression
+run_uuid: 0d7abd2ef2804a57b1754adba66adb00
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746436635202
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/metrics/accuracy b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/metrics/accuracy
new file mode 100644
index 00000000..a9789ecb
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/metrics/accuracy
@@ -0,0 +1 @@
+1746436637248 0.9746376811594203 0
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/params/model_name b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/params/model_name
new file mode 100644
index 00000000..da45759e
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/params/model_name
@@ -0,0 +1 @@
+Logistic Regression
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.runName
new file mode 100644
index 00000000..da45759e
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.runName
@@ -0,0 +1 @@
+Logistic Regression
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/0d7abd2ef2804a57b1754adba66adb00/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/meta.yaml b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/meta.yaml
new file mode 100644
index 00000000..15f7e2b7
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/artifacts
+end_time: 1746436662299
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 344892be3f9046c6aac8f2ef793402bf
+run_name: Naive Bayes
+run_uuid: 344892be3f9046c6aac8f2ef793402bf
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746436661807
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/metrics/accuracy b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/metrics/accuracy
new file mode 100644
index 00000000..4b69be84
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/metrics/accuracy
@@ -0,0 +1 @@
+1746436662256 0.9692028985507246 0
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/params/model_name b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/params/model_name
new file mode 100644
index 00000000..a7fb5317
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/params/model_name
@@ -0,0 +1 @@
+Naive Bayes
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.runName
new file mode 100644
index 00000000..a7fb5317
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.runName
@@ -0,0 +1 @@
+Naive Bayes
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/344892be3f9046c6aac8f2ef793402bf/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/meta.yaml b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/meta.yaml
new file mode 100644
index 00000000..2f9226a8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/artifacts
+end_time: null
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 439ee85dc07c435993ffeb0c7599c4dc
+run_name: traveling-shrike-717
+run_uuid: 439ee85dc07c435993ffeb0c7599c4dc
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746431087881
+status: 1
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/metrics/test_accuracy b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/metrics/test_accuracy
new file mode 100644
index 00000000..f458cd6e
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/metrics/test_accuracy
@@ -0,0 +1 @@
+1746431087953 0.975 0
diff --git a/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.runName
new file mode 100644
index 00000000..44f3d928
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.runName
@@ -0,0 +1 @@
+traveling-shrike-717
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/439ee85dc07c435993ffeb0c7599c4dc/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/MLmodel b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/MLmodel
new file mode 100644
index 00000000..0b8c615b
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/MLmodel
@@ -0,0 +1,21 @@
+artifact_path: Best_model
+flavors:
+  python_function:
+    env:
+      conda: conda.yaml
+      virtualenv: python_env.yaml
+    loader_module: mlflow.sklearn
+    model_path: model.pkl
+    predict_fn: predict
+    python_version: 3.12.4
+  sklearn:
+    code: null
+    pickled_model: model.pkl
+    serialization_format: cloudpickle
+    sklearn_version: 1.4.2
+mlflow_version: 2.21.3
+model_size_bytes: 73043
+model_uuid: 2d4dd98a11a7402f9a2b6016c555f3de
+prompts: null
+run_id: 4988088e75f54080af8a8be91668eaf9
+utc_time_created: '2025-05-05 09:17:44.253960'
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/conda.yaml b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/conda.yaml
new file mode 100644
index 00000000..3524d362
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/conda.yaml
@@ -0,0 +1,14 @@
+channels:
+- conda-forge
+dependencies:
+- python=3.12.4
+- pip<=25.1.1
+- pip:
+  - mlflow==2.21.3
+  - cloudpickle==2.2.1
+  - lz4==4.3.2
+  - numpy==1.26.4
+  - psutil==5.9.0
+  - scikit-learn==1.4.2
+  - scipy==1.13.1
+name: mlflow-env
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/model.pkl b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/model.pkl
new file mode 100644
index 00000000..d1d966d3
Binary files /dev/null and b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/model.pkl differ
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/python_env.yaml b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/python_env.yaml
new file mode 100644
index 00000000..61532bd5
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/python_env.yaml
@@ -0,0 +1,7 @@
+python: 3.12.4
+build_dependencies:
+- pip==25.1.1
+- setuptools==69.5.1
+- wheel==0.43.0
+dependencies:
+- -r requirements.txt
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/requirements.txt b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/requirements.txt
new file mode 100644
index 00000000..8c28dd75
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts/Best_model/requirements.txt
@@ -0,0 +1,7 @@
+mlflow==2.21.3
+cloudpickle==2.2.1
+lz4==4.3.2
+numpy==1.26.4
+psutil==5.9.0
+scikit-learn==1.4.2
+scipy==1.13.1
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/meta.yaml b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/meta.yaml
new file mode 100644
index 00000000..ea390ff4
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/artifacts
+end_time: 1746436692194
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 4988088e75f54080af8a8be91668eaf9
+run_name: thundering-newt-13
+run_uuid: 4988088e75f54080af8a8be91668eaf9
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746436664146
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.log-model.history b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.log-model.history
new file mode 100644
index 00000000..bfe44fd0
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.log-model.history
@@ -0,0 +1 @@
+[{"run_id": "4988088e75f54080af8a8be91668eaf9", "artifact_path": "Best_model", "utc_time_created": "2025-05-05 09:17:44.253960", "model_uuid": "2d4dd98a11a7402f9a2b6016c555f3de", "flavors": {"python_function": {"model_path": "model.pkl", "predict_fn": "predict", "loader_module": "mlflow.sklearn", "python_version": "3.12.4", "env": {"conda": "conda.yaml", "virtualenv": "python_env.yaml"}}, "sklearn": {"pickled_model": "model.pkl", "sklearn_version": "1.4.2", "serialization_format": "cloudpickle", "code": null}}}]
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.runName
new file mode 100644
index 00000000..236cfc96
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.runName
@@ -0,0 +1 @@
+thundering-newt-13
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4988088e75f54080af8a8be91668eaf9/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/meta.yaml b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/meta.yaml
new file mode 100644
index 00000000..1cf8950f
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/artifacts
+end_time: 1746436661783
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 4db50b0e36da4b558120acc28f9eca15
+run_name: Support Vector Classifier
+run_uuid: 4db50b0e36da4b558120acc28f9eca15
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746436644324
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/metrics/accuracy b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/metrics/accuracy
new file mode 100644
index 00000000..c3a70e09
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/metrics/accuracy
@@ -0,0 +1 @@
+1746436661722 0.9746376811594203 0
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/params/model_name b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/params/model_name
new file mode 100644
index 00000000..f7ff9246
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/params/model_name
@@ -0,0 +1 @@
+Support Vector Classifier
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.runName
new file mode 100644
index 00000000..f7ff9246
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.runName
@@ -0,0 +1 @@
+Support Vector Classifier
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/4db50b0e36da4b558120acc28f9eca15/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/meta.yaml b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/meta.yaml
new file mode 100644
index 00000000..422ec394
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/artifacts
+end_time: 1746430991253
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 70a859cda6854f51af32a22c6c4e789e
+run_name: Support Vector Classifier
+run_uuid: 70a859cda6854f51af32a22c6c4e789e
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746430985792
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/metrics/accuracy b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/metrics/accuracy
new file mode 100644
index 00000000..61330512
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/metrics/accuracy
@@ -0,0 +1 @@
+1746430991223 0.9746376811594203 0
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/params/model_name b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/params/model_name
new file mode 100644
index 00000000..f7ff9246
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/params/model_name
@@ -0,0 +1 @@
+Support Vector Classifier
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.runName
new file mode 100644
index 00000000..f7ff9246
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.runName
@@ -0,0 +1 @@
+Support Vector Classifier
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/70a859cda6854f51af32a22c6c4e789e/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/meta.yaml b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/meta.yaml
new file mode 100644
index 00000000..8c0f3a53
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/artifacts
+end_time: 1746430985788
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 75114b494eb94019897db9a04fd35abe
+run_name: Random Forest
+run_uuid: 75114b494eb94019897db9a04fd35abe
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746430983287
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/metrics/accuracy b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/metrics/accuracy
new file mode 100644
index 00000000..7aad923d
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/metrics/accuracy
@@ -0,0 +1 @@
+1746430985766 0.9637681159420289 0
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/params/model_name b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/params/model_name
new file mode 100644
index 00000000..53167014
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/params/model_name
@@ -0,0 +1 @@
+Random Forest
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.runName
new file mode 100644
index 00000000..53167014
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.runName
@@ -0,0 +1 @@
+Random Forest
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/75114b494eb94019897db9a04fd35abe/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/MLmodel b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/MLmodel
new file mode 100644
index 00000000..16efd7a9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/MLmodel
@@ -0,0 +1,21 @@
+artifact_path: Best_model
+flavors:
+  python_function:
+    env:
+      conda: conda.yaml
+      virtualenv: python_env.yaml
+    loader_module: mlflow.sklearn
+    model_path: model.pkl
+    predict_fn: predict
+    python_version: 3.12.4
+  sklearn:
+    code: null
+    pickled_model: model.pkl
+    serialization_format: cloudpickle
+    sklearn_version: 1.4.2
+mlflow_version: 2.21.3
+model_size_bytes: 73043
+model_uuid: d8eb0a115ae746eeaa35e70ad006d177
+prompts: null
+run_id: 78a22df00e284b7599bc7b9ff30f6e1a
+utc_time_created: '2025-05-05 07:44:36.871293'
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/conda.yaml b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/conda.yaml
new file mode 100644
index 00000000..3524d362
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/conda.yaml
@@ -0,0 +1,14 @@
+channels:
+- conda-forge
+dependencies:
+- python=3.12.4
+- pip<=25.1.1
+- pip:
+  - mlflow==2.21.3
+  - cloudpickle==2.2.1
+  - lz4==4.3.2
+  - numpy==1.26.4
+  - psutil==5.9.0
+  - scikit-learn==1.4.2
+  - scipy==1.13.1
+name: mlflow-env
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/model.pkl b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/model.pkl
new file mode 100644
index 00000000..d1d966d3
Binary files /dev/null and b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/model.pkl differ
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/python_env.yaml b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/python_env.yaml
new file mode 100644
index 00000000..61532bd5
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/python_env.yaml
@@ -0,0 +1,7 @@
+python: 3.12.4
+build_dependencies:
+- pip==25.1.1
+- setuptools==69.5.1
+- wheel==0.43.0
+dependencies:
+- -r requirements.txt
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/requirements.txt b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/requirements.txt
new file mode 100644
index 00000000..8c28dd75
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts/Best_model/requirements.txt
@@ -0,0 +1,7 @@
+mlflow==2.21.3
+cloudpickle==2.2.1
+lz4==4.3.2
+numpy==1.26.4
+psutil==5.9.0
+scikit-learn==1.4.2
+scipy==1.13.1
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/meta.yaml b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/meta.yaml
new file mode 100644
index 00000000..fe0990b8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/artifacts
+end_time: 1746431087844
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 78a22df00e284b7599bc7b9ff30f6e1a
+run_name: sincere-sow-936
+run_uuid: 78a22df00e284b7599bc7b9ff30f6e1a
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746431076820
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.log-model.history b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.log-model.history
new file mode 100644
index 00000000..c4a47e16
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.log-model.history
@@ -0,0 +1 @@
+[{"run_id": "78a22df00e284b7599bc7b9ff30f6e1a", "artifact_path": "Best_model", "utc_time_created": "2025-05-05 07:44:36.871293", "model_uuid": "d8eb0a115ae746eeaa35e70ad006d177", "flavors": {"python_function": {"model_path": "model.pkl", "predict_fn": "predict", "loader_module": "mlflow.sklearn", "python_version": "3.12.4", "env": {"conda": "conda.yaml", "virtualenv": "python_env.yaml"}}, "sklearn": {"pickled_model": "model.pkl", "sklearn_version": "1.4.2", "serialization_format": "cloudpickle", "code": null}}}]
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.runName
new file mode 100644
index 00000000..48a28efb
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.runName
@@ -0,0 +1 @@
+sincere-sow-936
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/78a22df00e284b7599bc7b9ff30f6e1a/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/meta.yaml b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/meta.yaml
new file mode 100644
index 00000000..c84ae481
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/artifacts
+end_time: 1746436644293
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 8250554e1c594fa09ff949b797401342
+run_name: Random Forest
+run_uuid: 8250554e1c594fa09ff949b797401342
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746436637410
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/metrics/accuracy b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/metrics/accuracy
new file mode 100644
index 00000000..9517393b
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/metrics/accuracy
@@ -0,0 +1 @@
+1746436644205 0.9637681159420289 0
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/params/model_name b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/params/model_name
new file mode 100644
index 00000000..53167014
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/params/model_name
@@ -0,0 +1 @@
+Random Forest
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.runName
new file mode 100644
index 00000000..53167014
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.runName
@@ -0,0 +1 @@
+Random Forest
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8250554e1c594fa09ff949b797401342/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/meta.yaml b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/meta.yaml
new file mode 100644
index 00000000..a80f4cd9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/artifacts
+end_time: 1746430983270
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 8a3286bbcc1949e1b0e7a359cd10c2cf
+run_name: Logistic Regression
+run_uuid: 8a3286bbcc1949e1b0e7a359cd10c2cf
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746430982592
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/metrics/accuracy b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/metrics/accuracy
new file mode 100644
index 00000000..300aa6e9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/metrics/accuracy
@@ -0,0 +1 @@
+1746430983238 0.9746376811594203 0
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/params/model_name b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/params/model_name
new file mode 100644
index 00000000..da45759e
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/params/model_name
@@ -0,0 +1 @@
+Logistic Regression
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.runName
new file mode 100644
index 00000000..da45759e
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.runName
@@ -0,0 +1 @@
+Logistic Regression
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/8a3286bbcc1949e1b0e7a359cd10c2cf/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/meta.yaml b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/meta.yaml
new file mode 100644
index 00000000..44148066
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/artifacts
+end_time: null
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: 9523c20a3b21442eae9fc67342b934ca
+run_name: fearless-frog-226
+run_uuid: 9523c20a3b21442eae9fc67342b934ca
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746436692246
+status: 1
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/metrics/test_accuracy b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/metrics/test_accuracy
new file mode 100644
index 00000000..0a5aa067
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/metrics/test_accuracy
@@ -0,0 +1 @@
+1746436692345 0.975 0
diff --git a/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.runName
new file mode 100644
index 00000000..75157dcc
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.runName
@@ -0,0 +1 @@
+fearless-frog-226
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/9523c20a3b21442eae9fc67342b934ca/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/meta.yaml b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/meta.yaml
new file mode 100644
index 00000000..bbe50a41
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/artifacts
+end_time: 1746429792909
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: d5d0308b5ae44800bbbfb115974eb266
+run_name: aged-grub-700
+run_uuid: d5d0308b5ae44800bbbfb115974eb266
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746429792833
+status: 4
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.runName
new file mode 100644
index 00000000..818694d7
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.runName
@@ -0,0 +1 @@
+aged-grub-700
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/d5d0308b5ae44800bbbfb115974eb266/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/meta.yaml b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/meta.yaml
new file mode 100644
index 00000000..96e3737f
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/meta.yaml
@@ -0,0 +1,15 @@
+artifact_uri: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/artifacts
+end_time: 1746430991557
+entry_point_name: ''
+experiment_id: '813981363149581026'
+lifecycle_stage: active
+run_id: dee67ae6eb2e4cc3bedd12ea155dc852
+run_name: Naive Bayes
+run_uuid: dee67ae6eb2e4cc3bedd12ea155dc852
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1746430991259
+status: 3
+tags: []
+user_id: kelly
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/metrics/accuracy b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/metrics/accuracy
new file mode 100644
index 00000000..63d0ad06
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/metrics/accuracy
@@ -0,0 +1 @@
+1746430991542 0.9692028985507246 0
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/params/model_name b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/params/model_name
new file mode 100644
index 00000000..a7fb5317
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/params/model_name
@@ -0,0 +1 @@
+Naive Bayes
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.runName b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.runName
new file mode 100644
index 00000000..a7fb5317
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.runName
@@ -0,0 +1 @@
+Naive Bayes
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.source.name b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.source.name
new file mode 100644
index 00000000..7b025773
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.source.name
@@ -0,0 +1 @@
+c:\Users\kelly\anaconda3\Lib\site-packages\ipykernel_launcher.py
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.source.type b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.source.type
new file mode 100644
index 00000000..0c2c1fe9
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.source.type
@@ -0,0 +1 @@
+LOCAL
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.user b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.user
new file mode 100644
index 00000000..4b9268c8
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/dee67ae6eb2e4cc3bedd12ea155dc852/tags/mlflow.user
@@ -0,0 +1 @@
+kelly
\ No newline at end of file
diff --git a/Streamlit/mlruns/813981363149581026/meta.yaml b/Streamlit/mlruns/813981363149581026/meta.yaml
new file mode 100644
index 00000000..93a3002d
--- /dev/null
+++ b/Streamlit/mlruns/813981363149581026/meta.yaml
@@ -0,0 +1,6 @@
+artifact_location: file:///c:/Users/kelly/Desktop/NLP%20Project-cloned/Explora_AI_Team_14_Classification_Project/Streamlit/mlruns/813981363149581026
+creation_time: 1746429792104
+experiment_id: '813981363149581026'
+last_update_time: 1746429792104
+lifecycle_stage: active
+name: News Category Classification
diff --git a/Streamlit/predictions.csv b/Streamlit/predictions.csv
new file mode 100644
index 00000000..8abc9cfd
--- /dev/null
+++ b/Streamlit/predictions.csv
@@ -0,0 +1,2001 @@
+category
+business
+education
+entertainment
+technology
+business
+technology
+technology
+sports
+education
+entertainment
+entertainment
+business
+education
+business
+technology
+technology
+business
+sports
+entertainment
+technology
+education
+sports
+business
+education
+sports
+entertainment
+technology
+education
+technology
+technology
+sports
+education
+entertainment
+business
+technology
+business
+entertainment
+entertainment
+business
+technology
+business
+entertainment
+sports
+business
+entertainment
+education
+education
+education
+technology
+education
+entertainment
+business
+education
+technology
+sports
+sports
+sports
+education
+sports
+education
+technology
+technology
+business
+education
+entertainment
+business
+sports
+technology
+sports
+business
+sports
+business
+entertainment
+education
+business
+technology
+sports
+entertainment
+business
+business
+entertainment
+sports
+sports
+technology
+education
+entertainment
+technology
+technology
+education
+entertainment
+entertainment
+sports
+business
+technology
+business
+technology
+education
+entertainment
+business
+technology
+business
+technology
+sports
+technology
+business
+entertainment
+education
+sports
+entertainment
+entertainment
+entertainment
+technology
+sports
+technology
+entertainment
+technology
+education
+business
+technology
+education
+sports
+business
+business
+technology
+business
+technology
+entertainment
+entertainment
+sports
+entertainment
+entertainment
+business
+business
+education
+sports
+technology
+education
+sports
+education
+technology
+entertainment
+entertainment
+business
+entertainment
+technology
+sports
+technology
+business
+sports
+education
+sports
+technology
+education
+education
+education
+sports
+education
+sports
+business
+sports
+business
+technology
+sports
+business
+education
+business
+education
+business
+business
+education
+sports
+technology
+education
+sports
+technology
+technology
+sports
+entertainment
+technology
+sports
+sports
+education
+entertainment
+sports
+entertainment
+sports
+education
+education
+sports
+technology
+sports
+sports
+education
+business
+education
+technology
+business
+technology
+technology
+business
+business
+entertainment
+technology
+entertainment
+technology
+technology
+entertainment
+business
+technology
+sports
+sports
+education
+education
+sports
+education
+business
+education
+sports
+sports
+business
+business
+business
+sports
+business
+sports
+business
+education
+entertainment
+sports
+sports
+entertainment
+sports
+business
+sports
+entertainment
+business
+entertainment
+education
+education
+business
+technology
+entertainment
+entertainment
+technology
+business
+technology
+entertainment
+business
+sports
+education
+education
+entertainment
+business
+education
+education
+business
+education
+education
+education
+technology
+sports
+education
+education
+sports
+technology
+education
+sports
+business
+business
+education
+sports
+education
+entertainment
+entertainment
+entertainment
+business
+technology
+sports
+education
+entertainment
+technology
+technology
+business
+entertainment
+technology
+entertainment
+sports
+entertainment
+business
+technology
+education
+education
+business
+technology
+entertainment
+sports
+sports
+business
+technology
+education
+business
+technology
+education
+sports
+business
+business
+sports
+sports
+technology
+sports
+technology
+business
+education
+business
+business
+technology
+entertainment
+entertainment
+business
+business
+education
+entertainment
+entertainment
+education
+technology
+education
+education
+technology
+sports
+business
+sports
+entertainment
+education
+education
+entertainment
+entertainment
+technology
+education
+sports
+technology
+sports
+sports
+sports
+sports
+business
+technology
+sports
+education
+entertainment
+entertainment
+education
+business
+business
+business
+sports
+entertainment
+sports
+sports
+education
+education
+education
+technology
+education
+education
+sports
+education
+entertainment
+entertainment
+sports
+sports
+education
+entertainment
+sports
+education
+entertainment
+entertainment
+business
+entertainment
+entertainment
+sports
+technology
+business
+entertainment
+technology
+education
+entertainment
+entertainment
+sports
+education
+entertainment
+entertainment
+technology
+education
+business
+entertainment
+technology
+technology
+entertainment
+technology
+technology
+sports
+education
+education
+education
+education
+entertainment
+technology
+entertainment
+sports
+technology
+entertainment
+business
+entertainment
+technology
+education
+business
+sports
+entertainment
+entertainment
+business
+technology
+education
+entertainment
+sports
+sports
+education
+business
+education
+sports
+entertainment
+education
+entertainment
+technology
+technology
+technology
+entertainment
+education
+sports
+technology
+entertainment
+entertainment
+business
+technology
+entertainment
+business
+education
+education
+sports
+technology
+sports
+entertainment
+business
+education
+business
+business
+business
+technology
+education
+business
+sports
+business
+sports
+education
+business
+sports
+education
+sports
+education
+education
+sports
+technology
+business
+entertainment
+education
+sports
+technology
+business
+business
+technology
+entertainment
+entertainment
+education
+sports
+business
+entertainment
+sports
+education
+technology
+technology
+education
+education
+sports
+education
+sports
+sports
+education
+sports
+sports
+entertainment
+sports
+technology
+sports
+entertainment
+sports
+business
+entertainment
+sports
+business
+entertainment
+technology
+sports
+business
+technology
+business
+entertainment
+technology
+entertainment
+entertainment
+entertainment
+sports
+sports
+education
+education
+entertainment
+technology
+entertainment
+education
+business
+business
+entertainment
+business
+sports
+business
+entertainment
+entertainment
+business
+education
+technology
+entertainment
+business
+entertainment
+technology
+education
+education
+education
+business
+entertainment
+technology
+business
+entertainment
+business
+technology
+technology
+sports
+sports
+sports
+entertainment
+education
+education
+sports
+business
+technology
+technology
+entertainment
+entertainment
+education
+business
+business
+technology
+entertainment
+sports
+technology
+sports
+sports
+technology
+business
+technology
+entertainment
+education
+sports
+business
+sports
+education
+business
+sports
+entertainment
+sports
+entertainment
+education
+entertainment
+business
+education
+business
+sports
+entertainment
+business
+education
+entertainment
+sports
+sports
+education
+technology
+business
+technology
+education
+business
+sports
+education
+education
+sports
+entertainment
+technology
+technology
+technology
+technology
+business
+sports
+entertainment
+business
+business
+entertainment
+technology
+technology
+sports
+sports
+business
+entertainment
+technology
+business
+technology
+sports
+technology
+education
+technology
+entertainment
+education
+entertainment
+education
+education
+business
+sports
+technology
+entertainment
+sports
+education
+sports
+business
+sports
+entertainment
+education
+education
+business
+education
+technology
+entertainment
+entertainment
+entertainment
+entertainment
+technology
+business
+entertainment
+entertainment
+technology
+education
+technology
+entertainment
+business
+education
+sports
+education
+entertainment
+entertainment
+sports
+business
+technology
+business
+sports
+technology
+technology
+education
+entertainment
+sports
+education
+sports
+sports
+education
+business
+entertainment
+technology
+business
+technology
+education
+technology
+technology
+technology
+education
+sports
+technology
+entertainment
+business
+education
+business
+technology
+entertainment
+business
+education
+business
+education
+technology
+sports
+education
+education
+business
+technology
+technology
+sports
+education
+sports
+education
+sports
+sports
+sports
+education
+business
+business
+education
+technology
+business
+sports
+business
+entertainment
+sports
+education
+education
+education
+technology
+sports
+entertainment
+education
+technology
+entertainment
+business
+technology
+business
+education
+business
+business
+education
+education
+sports
+technology
+entertainment
+business
+sports
+entertainment
+education
+entertainment
+sports
+business
+technology
+sports
+business
+business
+education
+business
+sports
+technology
+entertainment
+business
+education
+entertainment
+education
+technology
+business
+technology
+sports
+business
+sports
+business
+sports
+technology
+sports
+sports
+technology
+business
+education
+entertainment
+sports
+business
+sports
+business
+entertainment
+entertainment
+business
+education
+entertainment
+business
+education
+business
+entertainment
+sports
+sports
+sports
+business
+technology
+business
+education
+education
+technology
+entertainment
+business
+technology
+entertainment
+technology
+education
+business
+business
+entertainment
+sports
+technology
+business
+entertainment
+education
+education
+sports
+business
+education
+sports
+education
+education
+technology
+education
+business
+sports
+education
+technology
+technology
+sports
+technology
+business
+business
+business
+education
+sports
+business
+sports
+technology
+business
+technology
+technology
+business
+education
+entertainment
+technology
+business
+sports
+technology
+sports
+technology
+entertainment
+education
+sports
+entertainment
+entertainment
+entertainment
+entertainment
+business
+sports
+entertainment
+sports
+business
+sports
+technology
+entertainment
+business
+education
+education
+entertainment
+entertainment
+sports
+entertainment
+entertainment
+technology
+sports
+education
+business
+entertainment
+technology
+sports
+sports
+entertainment
+entertainment
+sports
+entertainment
+sports
+business
+sports
+technology
+sports
+entertainment
+technology
+entertainment
+entertainment
+entertainment
+business
+sports
+sports
+sports
+entertainment
+education
+technology
+business
+business
+entertainment
+technology
+technology
+education
+technology
+business
+technology
+business
+sports
+entertainment
+sports
+technology
+business
+business
+technology
+entertainment
+education
+sports
+technology
+education
+business
+entertainment
+technology
+entertainment
+business
+education
+technology
+sports
+technology
+business
+business
+entertainment
+technology
+sports
+entertainment
+education
+education
+education
+entertainment
+education
+business
+business
+sports
+sports
+business
+business
+education
+sports
+sports
+business
+sports
+technology
+entertainment
+sports
+technology
+technology
+technology
+technology
+sports
+education
+education
+business
+entertainment
+business
+business
+entertainment
+sports
+entertainment
+technology
+entertainment
+business
+sports
+sports
+education
+business
+technology
+entertainment
+business
+sports
+education
+sports
+sports
+technology
+business
+entertainment
+business
+business
+technology
+sports
+business
+sports
+business
+education
+sports
+technology
+sports
+education
+sports
+technology
+technology
+entertainment
+sports
+business
+business
+entertainment
+entertainment
+sports
+business
+technology
+technology
+business
+sports
+education
+sports
+education
+sports
+technology
+business
+entertainment
+sports
+technology
+sports
+sports
+sports
+business
+technology
+entertainment
+entertainment
+education
+technology
+entertainment
+business
+education
+sports
+technology
+technology
+business
+technology
+entertainment
+sports
+entertainment
+entertainment
+business
+sports
+technology
+entertainment
+education
+entertainment
+entertainment
+technology
+sports
+technology
+business
+sports
+technology
+technology
+sports
+sports
+sports
+entertainment
+entertainment
+technology
+technology
+education
+education
+business
+entertainment
+education
+business
+education
+technology
+sports
+education
+education
+entertainment
+education
+technology
+business
+education
+entertainment
+business
+education
+business
+sports
+sports
+sports
+entertainment
+technology
+business
+sports
+business
+education
+technology
+education
+entertainment
+sports
+business
+sports
+technology
+business
+business
+business
+sports
+sports
+business
+technology
+business
+entertainment
+technology
+entertainment
+education
+education
+education
+sports
+entertainment
+sports
+education
+entertainment
+sports
+education
+education
+entertainment
+sports
+entertainment
+entertainment
+sports
+business
+sports
+technology
+entertainment
+business
+technology
+business
+business
+entertainment
+sports
+business
+technology
+technology
+business
+entertainment
+education
+entertainment
+education
+entertainment
+entertainment
+entertainment
+education
+education
+business
+business
+education
+education
+business
+technology
+entertainment
+technology
+technology
+entertainment
+technology
+business
+education
+sports
+education
+sports
+sports
+sports
+education
+entertainment
+technology
+business
+technology
+entertainment
+sports
+entertainment
+sports
+entertainment
+technology
+sports
+business
+sports
+technology
+entertainment
+entertainment
+entertainment
+education
+sports
+sports
+technology
+business
+entertainment
+technology
+business
+business
+sports
+education
+education
+business
+sports
+technology
+education
+technology
+education
+entertainment
+entertainment
+sports
+technology
+technology
+sports
+sports
+business
+entertainment
+business
+education
+education
+business
+education
+education
+business
+entertainment
+education
+sports
+education
+sports
+business
+entertainment
+technology
+education
+entertainment
+sports
+entertainment
+technology
+education
+education
+entertainment
+sports
+technology
+entertainment
+sports
+technology
+entertainment
+entertainment
+technology
+education
+sports
+business
+technology
+technology
+entertainment
+business
+business
+entertainment
+business
+technology
+entertainment
+entertainment
+business
+entertainment
+business
+technology
+entertainment
+education
+education
+entertainment
+technology
+sports
+education
+business
+education
+sports
+technology
+business
+sports
+entertainment
+business
+education
+technology
+education
+sports
+sports
+business
+sports
+education
+business
+technology
+technology
+technology
+technology
+education
+education
+technology
+business
+sports
+education
+technology
+sports
+sports
+entertainment
+education
+business
+technology
+education
+sports
+sports
+technology
+technology
+technology
+business
+entertainment
+business
+sports
+business
+technology
+entertainment
+business
+business
+technology
+sports
+education
+education
+sports
+technology
+sports
+entertainment
+technology
+entertainment
+education
+technology
+entertainment
+business
+sports
+education
+technology
+technology
+business
+technology
+business
+education
+entertainment
+sports
+business
+entertainment
+entertainment
+technology
+sports
+sports
+sports
+entertainment
+entertainment
+education
+education
+sports
+technology
+technology
+technology
+education
+sports
+education
+entertainment
+business
+sports
+sports
+entertainment
+business
+education
+technology
+entertainment
+technology
+business
+technology
+entertainment
+sports
+business
+technology
+education
+entertainment
+entertainment
+entertainment
+education
+sports
+education
+entertainment
+technology
+education
+technology
+entertainment
+sports
+entertainment
+business
+technology
+entertainment
+entertainment
+technology
+entertainment
+sports
+business
+entertainment
+education
+education
+entertainment
+business
+technology
+business
+sports
+business
+entertainment
+entertainment
+education
+sports
+education
+education
+education
+sports
+entertainment
+sports
+education
+technology
+entertainment
+sports
+education
+education
+education
+business
+sports
+entertainment
+technology
+entertainment
+education
+technology
+technology
+technology
+business
+business
+technology
+education
+sports
+entertainment
+business
+entertainment
+business
+business
+technology
+education
+technology
+business
+sports
+education
+education
+education
+education
+education
+business
+entertainment
+business
+sports
+entertainment
+education
+sports
+technology
+technology
+entertainment
+sports
+sports
+education
+business
+education
+business
+business
+technology
+entertainment
+technology
+entertainment
+education
+education
+entertainment
+business
+technology
+business
+technology
+sports
+education
+technology
+education
+sports
+business
+business
+education
+business
+business
+business
+technology
+education
+technology
+education
+education
+technology
+sports
+entertainment
+entertainment
+entertainment
+sports
+education
+sports
+sports
+entertainment
+sports
+education
+business
+business
+technology
+entertainment
+business
+technology
+entertainment
+sports
+technology
+technology
+technology
+education
+business
+technology
+entertainment
+business
+technology
+technology
+technology
+education
+technology
+entertainment
+technology
+business
+education
+business
+entertainment
+entertainment
+business
+education
+entertainment
+education
+sports
+education
+sports
+entertainment
+entertainment
+entertainment
+business
+education
+education
+entertainment
+entertainment
+technology
+education
+technology
+technology
+technology
+entertainment
+entertainment
+business
+sports
+business
+technology
+business
+education
+sports
+business
+business
+sports
+technology
+education
+entertainment
+education
+sports
+entertainment
+business
+education
+education
+entertainment
+sports
+business
+entertainment
+business
+technology
+business
+sports
+business
+education
+technology
+education
+technology
+education
+entertainment
+sports
+education
+technology
+education
+business
+entertainment
+education
+sports
+entertainment
+sports
+entertainment
+sports
+sports
+education
+entertainment
+education
+entertainment
+sports
+technology
+technology
+sports
+technology
+technology
+technology
+entertainment
+education
+technology
+business
+technology
+education
+technology
+education
+education
+entertainment
+business
+technology
+education
+education
+sports
+technology
+sports
+business
+business
+technology
+technology
+sports
+business
+business
+entertainment
+education
+entertainment
+technology
+entertainment
+business
+entertainment
+technology
+technology
+business
+sports
+entertainment
+education
+technology
+business
+entertainment
+education
+education
+business
+technology
+business
+technology
+education
+technology
+entertainment
+sports
+entertainment
+business
+education
+education
+education
+business
+education
+business
+technology
+sports
+entertainment
+entertainment
+entertainment
+entertainment
+sports
+sports
+technology
+entertainment
+technology
+sports
+sports
+education
+entertainment
+technology
+technology
+education
+entertainment
+entertainment
+technology
+education
+entertainment
+sports
+technology
+business
+entertainment
+technology
+sports
+business
+education
+business
+sports
+technology
+education
+technology
+education
+technology
+technology
+sports
+business
+business
+technology
+education
+business
+sports
+entertainment
+entertainment
+business
+sports
+entertainment
+sports
+business
+sports
+education
+education
+entertainment
+education
+technology
+business
+education
+business
+business
+entertainment
+technology
+sports
+technology
+education
+education
+education
+education
+education
+business
+technology
+entertainment
+entertainment
+business
+business
+business
+technology
+business
+education
+education
+sports
+business
+education
+education
+education
+entertainment
+entertainment
+sports
+education
+business
+business
+education
+business
+education
+technology
+technology
+technology
+technology
+business
+sports
+sports
+technology
+technology
+sports
+entertainment
+entertainment
+technology
+sports
+business
+technology
+sports
+technology
+education
+business
+education
+business
+technology
+technology
+sports
+sports
+business
+education
+technology
+sports
+education
+technology
+sports
+sports
+entertainment
+education
+sports
+technology
+business
+technology
+education
+entertainment
+technology
+business
+entertainment
+technology
+technology
+sports
+technology
+entertainment
+education
+education
+technology
+business
+business
+education
+business
+sports
+sports
+education
+education
+education
+business
+technology
+sports
+technology
+entertainment
+business
+entertainment
+technology
+business
+education
+entertainment
+sports
+sports
+education
+entertainment
+entertainment
+sports
+business
+education
+technology
+technology
+education
+sports
+education
+sports
+education
+business
+entertainment
+education
+sports
+technology
+education
+education
+business
+technology
+sports
+technology
+sports
+education
+business
+sports
+technology
+entertainment
+entertainment
+entertainment
+education
+sports
+entertainment
+entertainment
+entertainment
+sports
+entertainment
+entertainment
+education
+sports
+technology
+business
+business
+technology
+business
+entertainment
+sports
+education
+sports
+entertainment
+business
+entertainment
+sports
+sports
+entertainment
+sports
+technology
+technology
+sports
+sports
+education
+sports
+technology
+sports
+sports
+entertainment
+entertainment
+sports
+education
+entertainment
+technology
+technology
+education
+entertainment
+entertainment
+business
+business
+technology
+technology
+sports
+sports
+entertainment
+technology
+education
+technology
+business
+entertainment
+business
+technology
+technology
+business
+business
+sports
+entertainment
+entertainment
+entertainment
+technology
+technology
+education
+sports
+business
+business
+education
+technology
+technology
+entertainment
+sports
+education
+sports
+education
+education
+technology
+business
+entertainment
+technology
+sports
+sports
+sports
+technology
+entertainment
+education
+business
+technology
+technology
+business
+sports
+technology
+business
+education
+entertainment
+technology
+business
+business
+business
+education
+business
+business
+entertainment
+technology
+entertainment
diff --git a/Streamlit/vectorizer.pkl b/Streamlit/vectorizer.pkl
new file mode 100644
index 00000000..55be106f
Binary files /dev/null and b/Streamlit/vectorizer.pkl differ
diff --git a/classification presentation.pdf b/classification presentation.pdf
new file mode 100644
index 00000000..da984755
Binary files /dev/null and b/classification presentation.pdf differ