Efficient Mining of Frequent Item Sets in Heterogeneous Data

Association rule mining has recently become a popular area of research. The most expensive step of discovering association rules is to find so-called frequent item sets. The focus of this paper is efficient mining of frequent item sets when the input data contains categorical and quantitative attributes. We propose a new Apriori-like algorithm to solve this problem. The new algorithm, that we have called Gradual Apriori, generates about 30% less candidates than the traditional algorithm. More importantly, the new algorithm has much (several times) smaller memory requirements. The main disadvantage of Gradual Apriori is an increased number of iterations that needs to be performed. However, in many cases the new algorithm should still result in better overall performance.