CasaTrova

What is CasaTrova?

CasaTrova is an innovative project that combines AI/ML concepts with basic web development knowledge to predict house prices based on various aspects such as property location, total area, number of rooms, number of bathrooms, and more. By leveraging machine learning algorithms and analyzing historical data, CasaTrova provides accurate and reliable price estimates. The project showcases the seamless integration of AI/ML technology within a user-friendly web application, enabling users to input property details and receive instant predictions. CasaTrova empowers individuals in the real estate market by providing valuable insights and aiding decision-making processes related to property transactions.

Project Architecture

First we have downloaded the data set of house price from kaggle.com, we have downloaded the dataset of Bangalore city in India.

Using this data-set we will build a machine learning model.
While building the model we came accross some cool concepts of Data Science such as Data cleaning, Feature engineering, One hot encoding, Outlier detection, Dimensionality Reduction.

Once the model is built, we will export it to a pickle file and then we will wright a python flask server which can consume this pickle file and do price prediction for us.
This python flask server will expose http end points for various requests and the UI return in HTML CSS and JS will make Http get and post calls.

Toots and technologies:

Python as programing language
Pandas for data cleaning
Matplotlib for data visualization
Sklearn for model building
Python flask for backend server
HTML, CSS and JS for our website.

CasaTrova

Figure-1
This is the layout of CasaTrova.

Here we need to select our desired location and along with BHK size, Number of bathroom and Total area.

Figure-2

Filling all the basic details.

In this case we have chose
5th phase JP Nage as our desired location
2BHK as our desired BHK size
2 as number of bathroom
1400 as total number of square feet
After filling all the basic details, we will hit predict price for the prediction.

Figure-3
Predicted Price

The predicted price of the selected e-state is shown.
In this case the predicted price is shown as 74.6 Lakh Rupees.