中药复方智能挖掘的关键技术研究及系统实现

发布时间：2018-05-29 20:36

本文选题：中医药学 + 复方分析　；参考：《西南交通大学》2017年硕士论文

【摘要】：中医药学,作为中华儿女千百年来通过不断实践、不断同各类疾病作斗争的优秀民族文化遗产,具备独特的诊疗理论体系,同时在历史发展的长河中也留下了许多传世的医学典著。目前,我国初步完成了该领域数据的信息化建设工作,人们可以通过这些中医药学数据共享平台检索自己所需的信息,但中医药学极具多样性和复杂性,且仅仅对着庞大的数据集进行检索势必不能很好地发挥其价值,需要通过结合现代数据分析手段去探索其中的奥秘。中药复方规律研究作为中医药学现代化的重要组成部分,不但丰富和发展了复方学理论,同时为复方临床应用和新药研发提供客观依据。本课题就中医药领域的配方规律展开研究,并设计了一个中医药学数据挖掘系统,为新药研发、古方探索等工作提供客观的参考。具体的研究内容如下:(1)介绍了中医药学相关数据的特点并分析了其中存在的数据规范化问题,对中药药物数据和复方数据的具体预处理手段进行详细的说明。(2)提出一种复方关联规则分析方法。针对传统关联规则挖掘算法中存在的不足,提出挖掘正负关联规则的方法,并针对于引入负事务项后存在的一些问题比如爆炸式频繁项集增长问题和算法的挖掘效率问题提出相应的解决方案。(3)提出一种新的复方网络化分析方法。对传统的中医药学复方数据进行网络化建模,然后对复方网络特性进行分析并以此为基础利用社团发现方法进行进一步的挖掘。(4)设计并实现了中医药学数据挖掘系统,该系统集成了数据预处理、复方关联规则分析、复方网络化分析等相关技术,为相关使用人员通过方便、简洁的交互接口。
[Abstract]:Chinese medicine, as an excellent national cultural heritage in which the Chinese people have been constantly fighting various diseases for thousands of years through continuous practice, has a unique theoretical system for diagnosis and treatment. At the same time, in the history of the long river also left a lot of ancient medical classics. At present, our country has preliminarily completed the information construction of the data in this field. People can retrieve the information they need through these data sharing platforms of traditional Chinese medicine, but Chinese medicine is very diverse and complex. It is necessary to explore the mystery by combining the modern data analysis with the search of large data sets which is bound to be unable to give full play to its value. As an important part of the modernization of traditional Chinese medicine, the study on the law of compound Chinese medicine not only enriches and develops the theory of compound medicine, but also provides an objective basis for the clinical application of compound medicine and the research and development of new drugs. In this paper, the formula law in the field of traditional Chinese medicine is studied, and a data mining system of traditional Chinese medicine is designed, which provides an objective reference for the research and development of new drugs and the exploration of ancient Chinese medicine. The specific contents of the study are as follows: (1) introducing the characteristics of the relevant data of Chinese medicine and analyzing the problems of data standardization in them. In this paper, we give a detailed description of the pretreatment methods of traditional Chinese medicine (TCM) data and compound data. (2) A compound association rule analysis method is proposed. Aiming at the shortcomings of traditional association rules mining algorithm, a method of mining positive and negative association rules is proposed. A new method of compound network analysis is proposed to solve the problem of explosive frequent itemsets growth and mining efficiency of the algorithm. Based on the network modeling of traditional Chinese medicine compound data, this paper analyzes the characteristics of the compound network and uses the community discovery method to further mine the data mining system of traditional Chinese medicine and implements the data mining system of traditional Chinese medicine. The system integrates data preprocessing, complex association rule analysis, compound network analysis and other related technologies, through convenient and concise interactive interface for relevant users.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R2-03;TP311.13

【参考文献】