Real Estate Prediction Using Deep Learning

Inspiration

Performed a similar project for a class using decision trees, this term I'm taking a Deep Learning course and decided to apply it to the same data to see how it improves the prediction accuracy

What it does

It has to be trained using data obtained from several datasets available on the Open Data Portal of the City of Edmonton. After proper data preprocessing and training the model should be able to predict the price of a new property with reasonable accuracy.

How I built it

I formatted the data using Microsoft Excel and MATLAB. The model was programmed using TensorFlow on Python.

Challenges I ran into

It is very hard to get the datasets to be compatible with what the TensorFlow code needs. The bugs in the system and also the size of the data make it hard to experiment and correct errors to reach a working solution. Using Python to program the model requires high precision and very clear data, any difference between the types of data will produce errors.

Accomplishments that I'm proud of

Deep learning is still developing and it is very new to me so any advancement and experience I get doing a project on it is considered a milestone for me, so even though I wasn't able to finish the project, I'm glad I got this far with it.

What I learned

I learned about many challenges that are faced in Deep Learning and how to solve the errors that come up regularly while using TensorFlow.

What's next for Real Estate Prediction Using Deep Learning

I will continue working on the model and make a phone app that can do the prediction task

Built With

• python
• tensorflow