Font Size: a A A

Research And Design Expert System For Diabetes Diagnosis Based On Decision Tree

Posted on:2017-02-08Degree:MasterType:Thesis
Country:ChinaCandidate:J Y NiFull Text:PDF
GTID:2334330488498674Subject:Computer technology
Abstract/Summary:PDF Full Text Request
With the development of artificial intelligence, the applications of artificial intelligence in medicine are more and more widespread. Expert system is an important branch of artificial intelligence applications, which provides an effective means to solve medical problems. Decision tree is an important model of expert system, it can efficiently express the knowledge with strong reasoning ability. This paper develops a diabetes diagnosis expert system for Ma'anshan Taoyuan community hospital based on the decision tree. The system is used for the diagnosis and treatment of diabetes in community residents.The main research contents of this paper are as follows:First of all, this paper introduces the background and significance of the subject and the research status at home and abroad, points out the shortcomings and deficiencies of several common expert system development technologies, proposes using decision tree to develop expert system.Secondly, this paper extracts the diabetes attributes by analysising the pathogenesis of diabetes, and processes the attributes including removing unnecessary attributes, discreting attributes, supplementing vacancy attributes. On the basis of decision tree learning, the decision tree model is built. After experiments, the validity of decision tree model to diagnosis diabetes is verified.Then, the structure, components, reasoning mechanism, reasoning direction of the expert system and the storage form of decision tree are designed. The relevant technologies used are introduced too.Finally, this paper uses the software engineering to design the system, which implements the function of diabetes diagnosising, case loading, knowledge base management and information management. The system is established by C/S structure.The interface of the system is developed by using Eclipse and Java Swing, knowledge base and comprehensive database are developed by using SQL Server 2008. The system has being installed and used to Ma'anshan Taoyuan community hospital for medical personnel.
Keywords/Search Tags:Decision Tree, Knowledge Representation, Expert System, Diabetes
PDF Full Text Request
Related items