Ishikawa7 / Simple-Collaborative-Filtering-with-Pandas

A basic example of collaborative filtering in Python using the MovieLens 100k dataset

Simple Collaborative Filtering

A basic example of collaborative filtering in Python using the MovieLens 100k dataset, the dataset contains ratings data for 9,724 movies by 610 users.

Requirements

pandas

Usage

To use the code, simply run the script in a Python interpreter.

Collaborative Filtering

Collaborative filtering is a method of making recommendations based on the ratings of similar users. In this example, i used Pearson correlation as the similarity measure and select the k nearest neighbors to make recommendations.

References

MovieLens 100k dataset: https://grouplens.org/datasets/movielens/100k/