Mining Weighted Rare Association Rules Using Sliding Window over Data Streams

Rare association rules mining is an association rule which has low support and high confidence. In recent years, the problem of mining rare association rules has got quite a lot of attention, which has become a hot topic in data mining research. However, most of the research on mining rare association rules are confined to the static database environment, and treat each item with the same significance although different items may have different significance. In this paper, we propose an algorithm for mining weighted rare association rules over data streams with a sliding window. Experiments on the synthetic data stream show that the proposed algorithm is efficient and scalable.