Constructing data mining model of five viscera correlation theory of Myasthenia Gravis based on rough set and association rules

Abstract

The 21st century is the era of digital information, and the problems with mathematical expressions have become common requirements for almost all subjects. The TCM five-viscera correlation theory is totally novel in theory. It thus needs to be presented mathematically. The relationship between the five viscera is explored using data mining algorithms. METHOD: 567 patients with Myasthenia Gravis (MG) were collected, and the clinical information of Chinese medical four diagnostic methods was recorded. The information was exploited to obtain the core symptoms of each organ using the modified attribute set reduction algorithm in the Rough Set theory. It was also used to explore the relationship of the five viscera using FP-tree algorithm in the Association Rules theory. RESULT: The core symptoms for each organ of MG are accessed. The bidirectional relationship between two viscera, the relationship between spleen and the other viscera, and the relationship among three viscera are achieved. CONCLUSION: The results are achieved by using algorithms in Rough Set and Association Rules theories conform to the clinical results. It shows that both methods are feasible in the mathematical study of the TCM five-viscera correlation theory.