Language used : python
Tools : Jupyter Notebook
Work flow :
- Collect preferences (ratings given by users to specific movies).
- Define similarity function to find similarity between two users.
- Here we are using pearson corelation coefficient as similarity metric
- Define top matches function to fetch n number of most similar user/items.
- Generate recommendations for a given user using defined similarity metric using weighted sum ranking.
- Flip the preferences from (ratings given to different movies by specific users) to (ratings given by different users to specific movie)
- Get preferences for given movie.
It's a hello world project of me in python and data science project. So please raise issues to improve it. Discussions, Contributions are always welcome.