Item/Product Recommender

Inspiration

We wanted to build an app that would help us survive the apocalypse.

What it does

This app is a machine learning script that takes any dataset which features products, users who reviewed them, and what rating they gave them. Based on this data, we are able to predict what products a user would be interested in based on what products they already like.

How we built it

We used a Python script which uses various machine learning techniques. One of those techniques is a Pearson Correlation formula. We also used other statistical techniques to analyze the data and make sense of it. All of this can be viewed in a python notebook, which we have as a html file in our Github directory.

Challenges we ran into

We had to make sense of the data, and try various machine learning techniques to see which would be the best. Eventually, we settled on the Pearson Correlation Formula.

What we learned

How to take advantage of the various machine learning techniques we had heard about.

What's next for Item Recommender

It can be expanded to be a fully featured web application, where people can get recommendations about specific products based on what other people think.