Reel Review

Reel Review
Page One
Page Two

Inspiration

Often times we find ourselves scrolling through Netflix unable to find an entertaining movie to watch. Thus, we were inspired to create Reel Review.

What it does

The application allows users to input movie reviews for any movies they have watched. The application uses this information to compute if the review was positive or negative. Other users can then look up movies, and discover how many people have stated that the movie was good, how many people have stated that the movie was bad, and the positive review percentage.

How we built it

To build the application, We made and trained a Tensorflow model to pick up patterns on positive movie reviews and negative movie reviews. We also used python and MongoDB to store and retrieve stored movie data. Using Tkinter, we created an intricate GUI as a median for the application.

Challenges we ran into

A Tensorflow model takes time to train, Resulting in long startup times. To fix this issue, we trained and saved a Tensorflow model, where it can then be quickly loaded at startup. Also, meshing our different languages together was hard to do.

Accomplishments that we're proud of

We are proud of creating a functional application which utilizes Tensorflow and MongoDB with a GUI in the span of a day.

What we learned

From this project, we learnt the fundamental features of MongoDB as well as efficient usage of resources. We were able to gain more knowledge about all the resources that are available online, and how it could help us learn more.

What's next for Reel Review

Next steps for reel review would involve a user sign in system, allowing for per user movie data. We also thought about implementing a "suggestion" box, in which movies that aren't in the system could be added. One thing that could be added down the line could be "community hubs", where people who enjoy the same series or genres are able to communicate.