MovieLens Dataset Movie Recommendation Dataset

The MovieLens dataset was released by the GroupLens group of the Department of Computer Science and Engineering at the University of Minnesota in 1998. This data can be used to train recommendation algorithms such as collaborative filtering, content-based recommendations, etc. Researchers can use this data to evaluate the performance of recommendation systems, conduct user behavior analysis, and develop new recommendation techniques.

MovieLens is a movie recommendation dataset. The information comes from the website movielens.umn.edu. The dataset has multiple versions, including but not limited to MovieLens 100K, MovieLens 1M, MovieLens 10M, MovieLens 20M, etc. Each version has different dataset sizes and the number of records it contains.

This website provides downloads of MovieLens 1M and MovieLens 20M datasets. The MovieLens 1M dataset contains 1 million ratings of 4,000 movies by 6,000 users, while the MovieLens 20M dataset contains 20 million ratings of 27,278 movies by 138,493 users, collected from January 1995 to March 2015.

The dataset contains the following files:

• ratings.dat: Contains the user's rating of a movie, usually in the format of UserID::MovieID::Rating::Timestamp.
• movies.dat: Contains the title and genre of the movie.
• users.dat: Contains basic information of the user, such as age, gender, occupation, etc.
• tags.dat: Tags given by users to movies.