ReviewsMark

• 

  Beta demo preview

Inspiration

We picked AI summarizing text track because we think it is challenged work and we came up with the idea of making something useful in the real world. We want to build a tool of summarizing Amazon reviews which will benefit both customers and sellers.

What it does

Given a URL of an Amazon product, our tool will summarize its positive and negative reviews made by customers. The outcome will be a list of pros and cons of the product.

How we built it

First, we use ´beautiful soup´ package to get all the reviews from the HTML file which contains all the information about the product we are analyzing.

Then, we train 2 word2vec models which are positive model and negative model respectively

Finally, we use K-means method to do clustering and find the centroids of clusters and the centroids are the representatives of the positive and negative reviews separately

Challenges we ran into

Considering our background, we figure out how to build an efficient model. Part of a team learned new technologies in a short time and how to adapt it to our solution.

Accomplishments that we're proud of

We are proud of having built a functional tool which could be used on any on-line shopping website.

What we learned

We learned flask. Got more experience at python programming language.

What's next for ReviewsMark

The next step would be to build a chrome extension which implements the same functionalities.

Built With

• flask
• python