RecDB is a recommendation engine built entirely inside PostgreSQL 9.2. It allows application developers to build recommendation applications using a wide variety of built-in recommendation algorithms such as user-user collaborative filtering, item-item collaborative filtering, and singular value decomposition. Applications powered by RecDB can produce online and flexible personalized recommendations to end-users. It is easily used and configured and allows novice developers to define a variety of recommenders that fits their application's needs in few lines of SQL. It can seamlessly integrate recommendation functionality with traditional database operations.
|Tags||recommender systems SQL Database postgresql postgres personalization machine learning data analysis collaborative filtering|
|Operating Systems||Linux Unix|
|Implementation||C++ Python Perl|
Release Notes: This release fixes User Filtering Delay in the Recommend Operator, optimizes memory consumption in user/item-based recommenders, and adds a comprehensive regression test suite.
Release Notes: This release adds "Recommend Query Plan" bugfixes.