Research On Symptom Standardization And Formula Composing Principle Of Traditional Chinese Medicine Based On The Firefly Algorithm

Chinese medicine is the carrier of the experience obtained when people struggled withdisease and the theoretical knowledge. In the process of Chinese medicine treatment, however,due to the difference of academic viewpoints and experiences in different places, times,physicians and objects, the writing of medical cases are always blur and not-standardized. In thecomputer assisted analysis of Chinese Medicine cases, the not-standardized medical cases are theobstacles of data mining. Thus, the standardized o f medical cases can not only eliminate theobstacles of computer assisted analysis but also improve the studies of medical cases. The coreprescription is the base treatment of a certain disease, according to the actual condition of thepatient, practitioners will get the prescription on the basis of the core prescription. Additional, inthe treatment of the same disease, something in different prescribes are in co mmon. In the largenumber of medical cases prescribed by famous doctors, the extraction of core prescription of thesame disease can help the interns grasp the treatment of similar disorders and shorten the trainingcycle. The problem of not-standardized medical case and how to seek prescription laws areconsidered in this paper, and we improve the firefly algorithm and embed it into the c-meansalgorithm to normalize medical cases and extract core prescription. The specific research resultsare as follows:(1) Even though firefly algorithm is an intelligent algorithm, its shortcomings are alsoapparent such as: the convergence precision is relatively low,the convergence rate is slow andtend to fall into local optimum. Taking this deficiencies into considered, a selection mechanismof combination of attraction and individual distance is proposed to avoid the problems discussedabove and improved the performance of the algorithm.(2) An improved firefly algorithm is embedded in the C-means clustering algorithm tooptimize the center of the FCM algorithm. An fuzzy clustering algorithm based on fireflyalgorithm is proposed in this paper, and it effectively overcome the shortcomings of FCM suchas sensitive initial values and tend to fall into local optimum. In the same time, it also improvedthe local search ability of firefly algorithm.(3) Automatic summarization base on the improved firefly clustering algorithm transformtheory point to a vector in the abstract model as the object to be extracted. We use a vector torepresent a sentence in the document, and obtain the standardized medical case after reorganize.(4) We apply the improved firefly clustering algorithm to manage the prescription ofChinese medicine, and FCM algorithm is applied to similar prescriptions to get the coreprescription. The extraction of core prescription provides a reference for learning Chinesemedical prescriptions, and a certain value for young doctors in pursuit of Chinese medicine.