Data Mining Apriori Algorithm Improvements And Applications In The Telecommunications Bi

Information technology is fruitful in the field of commercial applications. Enterprises can rapidly collect and processing business information through MIS (Management Information System), can accurately control the flow of information through ERP (enterprise resource planning system). These systems in addition to their application, but also have accumulated large amounts of data. Information system should have the ability to turn these huge data into knowledge, thereby supporting business decision-making, and even have automatically generate business decision-making ability, and this is Business Intelligence(BI). Information systems are experiencing the "MISâ†'ERPâ†'BI" process of evolution. From a global perspective, Business Intelligence (BI) has become the most promising field of information technology.Data mining as a high-level business application of BI is an essential part. Data Mining (DM) could find people interested knowledge from a large amount of data, it is a kind of deep-level data analysis methods. And is believed to be a effective method to resolve "data explosion of knowledge poverty". Association Rules Mining can better capture the important relationship between the data and found the rules simple easy-to-understand form. In recent years, Association Rules Mining has become a hot field of data mining.In this paper, begin with the introduction of the concept of the business intelligence and data mining technology, and a detailed discussion of the association rule mining technologies. Classical Apriori algorithm of association rules combined with transaction compression method and restriction rules improve the efficiency of the Apriori algorithm.Finally, the new algorithm is applied to data telecommunications business. Based on consumer information, we excavat more information for the marketing departments to support decision-making in order to promote the development of data services. At the same time, assess the algorithm from mining results, the experimental results are detailed analysis proves the advantages of the improved algorithm, and then the shortcomings of the new algorithm is pointed out.