多类数据中对照模式的挖掘

Identifying contrast pattern in multiple phenotypes are great challenge. Previous work only focuses on two groups. For example, we can compare the abnormal patients with normal patients to find patterns associated with disease. However, the simultaneous distinguish across multiple disease types has not been well studied yet, which is important for medical researcher. Association rule mining often generate a huge number of rules, but a bottleneck of them either are redundant or no sense in a specific domain. In this paper, we define VP (an acronym for "Vital Pattern") and PP (an acronym for "Protect Pattern") by a statistical metric, and propose a new algorithm called MVP to make use of the property discovery VP and PP from multiple phenotypes medical data, and draw a clearly causal graph. The algorithm can generate some useful rules for medical researchers. The results demonstrate the feasibility of performing the clinically useful classification from patients of multiple pneumonia types.