Association Rule Mining Method Based on Interest Measure with Positive and Negative Items

Negative item brings the increase of frequent items and makes association rules doubled. Interest measure is adopted to restrict the amount. By analyzing current interest measure models, the deviation-based interest measure is chosen and the confidence property is presented. An association rule mining algorithm based on interest measure with positive and negative items is described. And the analysis of conflict association rules is given. The experimental results indicate the given algorithm is efficient and feasible.