This app calculates the sale price of homes in USA. I used Scikit-Learn's Linear Regression algorithm to build the underlying model of this app. After trying various other ML models, it seems like in this situation a linear model works better than it's non linear counterparts. But before the model predicts the sale price of your house it normalizes the numeric inputs using Scikit-Learn's Standard Scaler and one-hot encodes the categorical inputs (text-based inputs). All these Scikit-Learn objects are packed into a model pipeline like the rest of the ML apps in my Webfolio. Here's an important thing to notice, the prices are calculated according to the USD money value in 1980 as mentioned in dataset card in Kaggle. So, you might have to adjust the prediction to the current year's USD value.
Area:- The total built up area of your house in Square Feet(sq.ft). Min value:- 1650, Max Value:- 16200
Storeys:- The number storeys in your house. Min value:- 1, Max Value:- 4
Bedrooms:- The bedroom count of your house. Min value:- 1, Max Value:- 6
Parking:- Number of parking spaces around and owned by this House. Min value:- 1, Max Value:- 6
Bathrooms:- Number of parking spaces around and owned by this House. Min value:- 1, Max Value:- 4
Basement:- This input determines whether your house has a basement. Available Options:- Yes, No
Prefarea:- This input determines whether your house is in a prime location. Available Options:- Yes, No
Hot Water Heating:- This input determines whether your house has hot water heating in its bathrooms. Available Options:- Yes, No
Air Conditioning:- This input determines whether your house has air conditioning in one or more rooms. Available Options:- Yes, No
Guest Room:- This input determines whether your house has a guest room. Available Options:- Yes, No
Furnishing Status:- This input determines whether your house has a guest room. Available Options:- Unfurnished, Semi - Furnished, Furnished