Research On Liver Cancer Risk Factors Based On Association Rules Method

In database technology matures, the rapid increase in the amount of data a variety of business today, data mining as a new set of database technology, artificial intelligence, machine learning, statistical and other interdisciplinary fields, decision-making behavior in all walks of life plays an increasingly important role. Data mining technology at finding from random, erratic, lacking a priori knowledge of massive data implied valuable knowledge to predict future trends advantages, good-looking to help people make decisions. Because of these advantages, making data mining technology has become one of the most popular computer science current direction, its economic value has emerged and is respected by many business organizations. The association rule mining is one of the most active data mining research method for improving the beginning of the birth of the decision-making process of the supermarket retail industry will play an important role. In medicine, the association rule analysis method used to find relationships between clinical symptoms and drugs, relationship between lifestyle and disease patients, hospital costs between disease and illness, so that sort of association rules, have made good results.This paper, from the concept of data warehouse are introduced and the construction started, gradually introducing the meaning of data mining, discusses and studies the association rules analysis of Apriori algorithm and the classical concepts and definitions, and for Apriori and its improved algorithm AprioriTid low efficiency and space occupancy rate less than the larger, presents a the improved algorithm based on AprioriTid, the algorithm to compress the things, project, in order to improve the efficiency of the space saving purposes. After a comparative analysis on the operation efficiency of the experiment, the conclusion is:the improved AprioriTid algorithm in mining massive data when improving the efficiency and space occupancy rate is relatively low, has good veracity and practicability. Finally, using Java programming language to achieve the optimized AprioriTid algorithm, data mining on13310cases of the First Affiliated Hospital of Guangxi traditional Chinese Medicine University of electronic medical records in the database data, the purpose is to find out the people living habits, environment, heredity caused some association rules between HCC students, to these rules as the analysis on the basis of many carcinogens, exploring the combination effects cause the possibility and risk of liver cancer, so as to obtain some relatively reliable liver cancer high risk early warning rule. These findings to improve the broad masses of the people, especially for those with high risk association rule conditions crowd, targeted for treatment and health care, regular physical examination, and medical personnel to provide the reference for the active early detection of cancer.