Discovery of Fuzzy Rare Association Rules from Large Transaction Databases

Rare association rules is an association rule which has low support and high confidence. In recent years, the discovery of rare association rules has got quite a lot of attention, which has become a hot topic in data mining research. However, current discovery algorithms for rare association rules are built on the binary valued transaction databases, which cannot deal with quantitative attributes. In this paper, we put forward a discovery algorithm for finding fuzzy rare association rules to handle quantitative attributes. Experiments on the synthetic data stream show that the proposed algorithm is efficient and scalable.