IWFPM: Interested Weighted Frequent Pattern Mining with Multiple Supports

Association rules mining has been under great attention and considered as one of momentous area in data mining. Classical association rules mining approaches make implicit assumption that items' importance is the same and set a single support for all items. This paper presents an efficient approach for mining users' interest weighted frequent patterns from a transactional database. Our paradigm is to assign appropriate minimum support (minsup) and weight for each item, which reduces the number of unnecessary patterns. Furthermore, we also extend the support-confidence framework and define an interest measure to the mining algorithm for excavating users' interested patterns effectively. In the end, experiments on both synthetic and real world datasets show that the proposed algorithm can generate more interested patterns.