/item-based-similarity

Implementation of item based similarity for MovieLens recommendations using hadoop to create the similarity matrix

Primary LanguageJava

item-based-similarity

Hadoop example for Montevideo BigData & DataScience Meetup. Implementation of item based similarity for MovieLens recommendations using hadoop to create the similarity matrix.

First Hadoop example

Simple hadoop job that creates a histogram of user ratings per movie. (edu.mapreduce.histogram.job.CountMovieUsersRatingsJob)

To run the job:

hadoop jar item-based-similarity-jar-with-dependencies.jar edu.mapreduce.histogram.job.CountMovieUsersRatingsJob <input> <output>

Provides Job unit tests using MRunit.

Custom Hadoop formats

Job that creates the history of movie ratings per user. Uses custom input and output formats to read and write the data. (edu.mapreduce.formats.job.UserMoviesRatingJob)

To run the job:

hadoop jar item-based-similarity-jar-with-dependencies.jar edu.mapreduce.formats.job.UserMoviesRatingJob <input> <output>