The Application And Research Of Decision Tree And Association Rules In The Analysis Of Drug Sale

Drug has become a necessity for daily life as a special commodity. With the reform of health care system and the development of medical, the fierce competition of drugstore is increasing. The traditional system of drug information management has been widely used in daily management of sales. But it is only able to provide simple statistics and quiry of records, and is not able to get hidden laws in the sales data. With the development of database technology, data mining as a new data processing and analysis techniques, has been widely used. Data mining is considered as a method of extraction, transformation, analysis and pattern processing, using that way to get useful information from the potentially database. Apply data mining technique to reveal the law of drug sales data, and then use these laws to assist the manager to adjust the strategy of sales. Thus, this will improve the ability of information management for medicine retail business.This thesis focuses on the sales data of drugstore, and researches the sales data by data mining techniques. Through modeling data, designing and implementing the appropriate algorithm, and apply the appropriate analysis method to the actual management system, thereby increasing the system function of the medical management system. The detailed contents of the thesis are as follows:(1) The model of decision tree is builded through the analysis of drug sales classification. Choose ID3 algorithm for the ananlysis algorithm through the experimental comparison, and improve ID3 algorithm. Do experiment for the analysis between the improved algorithm and the original algorithm. Decision tree is studied to analyze customer consumption and the frequency of member consumption. The improved ID3 algorithm is presented to analyze for giving a reasonable classification of the data.(2) Builds the associated model through the analysis of drugs. Analyze the correlation between drugs and categories of drugs by association rules. Base on the parameter sensitivity analysis, a reasonable parameter is designed, and then apply the Apriori algorithm to complete the analysis of correlation.(3) Design and implement the sales information management subsystem based on B/S structure, and embed the Java programming of the improved ID3 algorithm and the Apriori algorithm into the analysis module, achieve the appropriate analysis of data mining.