A Study Of Classification And Prediction Technology And Its Application In Data Mining Of Mobile Communications Business

As the competition and saturation are increasing in mobile communication market,the mobile communication company's marketing strategy has changed gradually from the previous product-centric theory to the present customer-centric theory. In order to improve the profits,the mobile communication company has to be customer-centric, understand customers deeply, guide customers, retain customers, enhance the value of existing customers,improve customer satisfaction and reduce customer loss rate.Effective customer retention through brand migration and effective complaints'handling through grading management of the complaining customers' credits is one of the main means currently used in the mobile communication companies.In this paper, with application in the marketing management demands such as the brand migration of New-Year-card customer and the credit rating of the complaining customers etc.,classification and prediction technology is studied for the application of dada mining and analysis in the mobile communication enterprises,highly practical algorithms and suitable models of data mining for mobile communication company's marketing and management's demand are also designed and developed, technical implementation proposal of the program is put forward which has achieved preferable application effects.The main job of this paper includes two points.Firstly, technology of data warehouse, data mining, business intelligence and its application in mobile communication enterprise is introduced and studied,and classification and prediction technology is especially stressed.Secondly, hidden Markov model method and analytic hierarchy process (AHP) method is studied with application in the engineering implementation of customer's brand migration and the evaluation and grading management of the complaining customers'credits.The innovative research of this paper is embodied not only in the arithmetic field but also in the engineering implementation.In order to solve the problems of highly computational complexity and subjectivity in the conventional AHP method,two novel methods are presented which are adjacent ratio vector method and information gain method.The methods have good effects in credit rating of the complaining customers of mobile communication corporation. In engineering implementation, classification and prediction technology based on hidden Markov model method is applied to brand migration in mobile communications companies,technical implementation of the program is put forward. The two kinds of the improved AHP method are applied to credit rating and grading management of the complaining customers in mobile communication companies, technical implementation of the program is also put forward which has achieved preferable application effects.