An Efficient Association Mining Method via Matrix Compression

Abstract

With the development of computer network technology and popularization, and the technology of acquisition, management and query develop rapidly. People who is in the surging data ocean, need an intelligent technology badly, to explore more valuable oil the data ocean,so the mining technology which has been successfully used in commercial, medical, financial, and other fields, emerges. The primary technology of mining is classification analysis, cluster analysis, Association Rule analysis, and etc. Association Rule analysis is an important branch of mining, which can extract the valuable associations or rules people want to know from warehouse. Based on the background of data platform for public petition, it aims to study how to combine the Association Rule analysis technology with the current massive government-data, extracting association information from the massive hidden government-data, which can provide userful and valuable information according to the practical significance of the associations for system managers and decision makers. This paper study the classic Association Rule algorithm Apriori, propose the improved Apriori algorithm based on matrix com-pression. Based on this improved Apriori algorithm, develop the module of the analysis of cases evaluation, which can associate the cases and case handlers according to the information of cases and case handlers fistly, and then get the practical significance from the association result analysis, which can provide help to the managers and decision makers.