Improving-Diversity-Performance-of-Association-Rule-Based-Recommender-Systems

Note that the dataset has been moved to movielense-experiment folder.

Order of running the files to run the code:

• src/movielensdataCleaner.py -- takes movielens data as input -- gives out frequent patterns and association rules (by confidence)
• src/main_1.py -- finds the dranks for all patterns constructed from association rules
• src/arrange_confidence_and_drank_ARs.py -- sorts both association rules based on confidence and drank
• src/make_recommendations.py -- takes input file as patterns for which to recommend -- gives out recommendations.