Self contained applications, called generators, are used to identify relevant relations between items and/or users which form the basis for generating recommendations. Currently easyrec comes with the following set of generators, each implementing a specific recommendation algorithm:
- Association Rule Miner: A shopping cart analyzer used to create recommendations of the form "users who bought X also bought Y".
- SlopeOne Recommender: A collaborative filtering algorithm using ratings for calculating recommendations.
- Profile Similarity Calculator (since v0.98): Calculates similarities between items based on properties stored in the profile system.
- Sample Generator: Generates N random recommendations for each item (as a sample for developers).
For a more detailed description of the background of these generators and how to adapt them to a specific situation, please follow visit the appropriate sections Association Rule Miner and SlopeOne Recommender .