Advance-house-price-prediction

Problem Statment

Using Different paramters related to house,we need to predict what will be the selling price of house

Approach

• First we checked the missing values
• We found the relationships betweeen missing values and target throgh bargraph, hence without dropping missing values we created label as missing in each missing valued categorical features
• For categorical features Target guided label encoding done to convert into numerical features

MOdel building

• We used Polynimial Linear Regression to predict the house price

Feature selection

• Select from model technique used to select the important features, Lasso regression used to select the important features

MOdel Performance

• The model is able to explain 91.2% variation in the data i.e r2_score is 91.2%

Model deployement

• Model is deployed in Local machine using Streamlit