Loan_Prediction

This is a simple loan approval prediction model using Dream Housing Finance Company dataset. This prediction takes various factors into consideration such as Gender, Marital Status, Education, Number of Dependents, Income, Loan Amount, Credit History and others through an online form.

Technologies:

• Programming Language: Python,HTML,CSS,Javascript
• Libraries: Pandas, Scikit-learn, Matplotlib, Numpy, Flask, Ajax

Approach

1. Filling missing values
2. Perform necessary cleaning and EDA.
3. One hot Encoding
4. Training various models and tune it.
5. Selecting most accurate model and exporting it in a pickle file(Using Pickle Library).
6. UI designing using HTML,CSS,Javascript
7. Build an webapp using Flask framework.

Model

• Logistic Regression Model
  • Accuracy : 81.54%
• SVM
  • Accuracy : 81.54%
• Random Forest
  • Accuracy : 75.12%
• Decision_Tree
  • Accuracy : 72.23%
• GaussianNB
  • Accuracy : 80.78%
• MultinomialNB
  • Accuracy : 69.86%

Demo

Data Source:

• Kaggle

Author

• Sumit Kothari