/Movie_Recommendation_System

A Python-based Program that recommends movies based on Emotion

Primary LanguagePythonApache License 2.0Apache-2.0

Movie_Recommendation_System

A Python-based Program that recommends movies based on Emotion

Basics Used:

  • Web Scraping

Recommendation:

  • Based on the emotion entered, each emotion has been quantified as follows:

  • Emotion || Recommended Genre

  • Sad – Drama

  • Disgust – Musical

  • Anger – Family

  • Anticipation – Thriller

  • Fear – Sport

  • Enjoyment – Thriller

  • Trust – Western

  • Surprise – Film-Noir

Scraping:

  • IMDb does not have an API , so data accumulation is performed by using scraping a particular URL of the accompanying emotion.
  • Based on the emotion entered, top 'n' movies for that emotion are displayed, where 'n' is entered by the user.

Files:

  • sentiment.py : actual program to scrape / get the movie recommendation
  • feels.py : a module in the package 'feeling', which takes in an emotion from sentiment, classifies it into one of 8 predefined emotions and returns said emotion category back to sentiment.py

Text files:

  • Each of the 8 emotions have a file with sample data inside them.
  • Sample Data for each file consists of words which lean towards / convey the particular emotion