Analysis of TCM hypertension syndrome types and core prescriptions based on SVD dimension reduction algorithms and association rules
Konferenz: BIBE 2019 - The Third International Conference on Biological Information and Biomedical Engineering
20.06.2019 - 22.06.2019 in Hangzhou, China
Tagungsband: BIBE 2019
Seiten: 7Sprache: EnglischTyp: PDFPersönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Lu, Gaole; Cao, Dong; Ye, Hui (School of Medical Information Engineering, Guangzhou University of Chinese Medicine, Panyu, Guangzhou, China)
In the prevention and treatment of hypertension, research shows that using Traditional Chinese Medicine (TCM) has the characteristics of high efficiency, sustainability and stability. There are a large number of data mining researches on hypertension medical records. However, the existing research on hypertension prescription mining is incomplete, some are lack of combination of syndrome analysis, some are clustering in high-dimensional sparse prescription matrix, which are in poor clustering effect. This paper will improve these points. For analyzing the association between composite syndrome types and clustering centres, we use SVD (Singular Value Decomposition) dimension reduction algorithm and k-means++ algorithm to get clustering centers and evaluate the clustering effect. We process the association rules by FP-growth algorithm. The results show that the SVD algorithm can be used to improve the clustering effect of prescriptions, and the seven core prescriptions can be matched to different complex syndrome types of hypertension.In addition, TCM physicians indicated that the conclusions in this paper is compatible of traditional Chinese medicine, and the medicine clusters, prescription are clear. The datamining results have practical significance and clinical feasibility.