Our team wanted to simplify the process of movie choice for everyone, starting from a huge movie fan to not a movie person. We understood that nowadays movies aren't categorized very well, as movie genres don't give an accurate picture of the plot.
What it does
We tried to split movies by emotions which they induce. We decided to make search by the most common emotions, which we analyzed by parsing IMDB reviews.
How I built it
We started from research and collecting data about different movies. As one of the most popular source we have chosen Imdb.com. Using python libs we have aggregated data from top movies. Then we trained SVM machine-learning algorithm with bag of words presentation data. Also we have tried to focus on user experience, so we built a web application with material design pattern.
Challenges I ran into
Elisa challenge "MY PERFECT PERSONAL MOVIE EXPERIENCE" link
Accomplishments that I'm proud of
We are proud of accuracy that we have achieved, using out algorithm. Also we think, that our application design is beautiful, easy and comfortable for users.
What I learned
We learned new libraries for NLP processing, preprocessing data, improved our knowledge in user experience and frontend.
What's next for Emotive
We are definitely convinced to continue our work. The next step for us is to improve recommendations, analyzing personal user movie ratings at Imdb.