Recommender
A C library for product recommendations/suggestions using collaborative filtering (CF).
Recommender analyzes the feedback of some users (implicit and explicit) and their preferences for some items. It learns patterns and predicts the most suitable products for a particular user.
Features
Collaborative Filtering
User and Item based recommenders
No external dependencies
Fast running time ~ 81 seconds for 10 million ratings (on MovieLens Data Sets)
Memory footprint under 160 MB for 10 million ratings
Webpage
Building
To compile Recommender:
make
The compilation will produce libRecommender.a
To compile an example:
gcc test/test.c src/libRecommender.a -lm -o test/t1 -I src/
Alternatively you can use clang
clang test/test.c src/libRecommender.a -lm -o test/t1 -I src/
Keywords
Collaborative filtering, recommender system
References