garvit-01/recommenders

Recommenders

Example Requirement: If someone liked the movie Contact (1997), we want to find out what would be the most similar movie to Contact (1997) to recommend to the user. The technique is being used is Pearson correlation.

Answer: is Philadelphia (1993) with a correlation coefficient of 0.446 with 137 ratings.

Please get the Data sets from below link before trying to run the recommender codes:

https://grouplens.org/datasets/movielens/

The codes are originally from:

https://towardsdatascience.com/how-to-build-a-simple-recommender-system-in-python-375093c3fb7d and/or https://www.analyticsvidhya.com/blog/2016/06/quick-guide-build-recommendation-engine-python/

As exercise, try other similarity techniques such as Jaccard and Cosine.