Study On Pretreatment Of TCM Clinical Data And Application Of Association Rules In Diabetes Mellitus

Diabetes is a serious disease, which is the third largest killer to human health. Nowadays, its etiology and pathogenesis still not clear in Western medicine, whose treatments and medicines is also lack. As shown by the clinical practice and research, Chinese medicines have a good effect in improving the symptoms of diabetes and prevent complications. Therefore, this study collected the clinical data which apply the traditional Chinese medicine treatment to the diabetes.This dissertation mainly introduced the establishment of term dictionary, applied the segmentation process on TCM clinical data, which contains a corresponding relation between standard and different nouns, and the processed data indicate whether the entry is a standard word, whether you need for entry in the process of data preprocessing specification provides the judgment basis, so as to improve the efficiency of data preprocessing, using the improved FP-growth algorithm for association rules mining data after pretreatment. Improved mainly in view of the structure in the process of FP-Tree, puts forward the idea of sharing the prefix, thereby reducing the construction process of FP-Tree medium voltage into a frequent item sets, children need to traverse the Tree nodes. The algorithm is a frequent itemsets generation algorithm, with the advantages of space and time, and does not produce a large number of candidate itemsets in the constituents of the operation.Combining with the professional characteristics and related literature, this papermain explore symptoms, signs and treatment on the Chinese medicine treatment, and analyze the clinical characteristics, to provide reference for clinical treatment of diabetes.