House Price Predictor Using Machine Learning

• 

  Result

• 

  Slope Determination

Inspiration

Due to the completely biased monopoly of Property Dealers in market, people migrating to modern cities have to face a crucial dilemma of buying a house at an optimal price. So our goal is to completely automate this whole property dealing sector where free will of people are given a primary concern.

What it does

We have designed an algorithm that trains the system through supervised learning methodology to predict the optimal price of a house based on certain criteria or features that the user might want.

How we built it

Instead of relying on superficial machine learning libraries, we have built the algorithm from scratch i.e. the Gradient Descent Algorithm. We basically train the system through a predefined dataset in order to let the system predict the optimal price.

Challenges we ran into

• Transforming the regression algorithm from linear to multivariate.
• Efficiency in performing mathematical calculations.
• Transforming conventional mathematical equations into vectorized form.

Accomplishments that we're proud of

• Successfully implemented the core engine of the whole application.
• Implemented every mathematical functions from scratch without using any libraries or APIs.

What we learned

• To implement Machine Learning concepts in real world.
• Concept of Automation.

What's next for House Price Predictor Using Machine Learning

• Here we have used only two features i.e. Area of the house and the number of Bedroom. But it can be expanded to evaluate the cost based on more number of features that might cover every aspect of housing and property dealing.
• At this stage the building of the core engine was more of a concern than how the software looks and feels. So, it can be deployed as a Web Application or Mobile Application.

Built With

• java