Mining Strongly Associated Rules

One of the main tasks of KDTCM (knowledge discovery in traditional Chinese medicine) is discovering novel paired or grouped drugs from Chinese medical formula database. Paired or grouped drugs, which are special combinations of two or more drugs, have strong efficacy. Association rule mining is used by reason of the large number of association relationships among various kinds of drugs. However, association rules reflect only one kind of association relationships and thus have less significance in TCM researches. In this paper, we propose to mine strongly associated rules, which have much more probability than association rules to be novel paired or grouped drugs because of strongly associated relationships between both sides of a rule. Experimental results on Chinese ancient medical formula database and traditional Chinese medicine herbal database show that all techniques developed in the paper are efficient and effective.