Analysis on traditional Chinese medicine prescriptions treating cancer based on traditional Chinese medicine inheritance assistance system and discovery of new prescriptions

Malignant tumor is one of the main causes for death in the world at present as well as a major disease seriously harming human health and life and restricting the social and economic development. There are many kinds of reports about traditional Chinese medicine patent prescriptions, empirical prescriptions and self-made prescriptions treating cancer, and prescription rules were often analyzed based on medication frequency. Such methods were applicable for discovering dominant experience but hard to have an innovative discovery and knowledge. In this paper, based on the traditional Chinese medicine inheritance assistance system, the software integration of mutual information improvement method, complex system entropy clustering and unsupervised entropy-level clustering data mining methods was adopted to analyze the rules of traditional Chinese medicine prescriptions for cancer. Totally 114 prescriptions were selected, the frequency of herbs in prescription was determined, and 85 core combinations and 13 new prescriptions were indentified. The traditional Chinese medicine inheritance assistance system, as a valuable traditional Chinese medicine research-supporting tool, can be used to record, manage, inquire and analyze prescription data.