Research On Clustering Methods Of Data Mining And Its Application

In recent years,data mining technique is a new science with the development of databaseand the artificial intelligence.As a new data extraction tool,it can convert the data into usefulinformation and knowledge automatically,deciders can make effective business strategies bythese information and knowledge.Just because data mining has such a bright businessprospects, it attracted wide concerns in academic and business community.Clustering analysis is a method of data mining.When we encounter massive data ,whatwe must do is classifying these data.The method of classifying the raw data is clusteringanalysis.Cluster analysis is widely used, in biology,plant and animal’s gene classification canbe derived by clustering analysis.In business market,clustering analysis can help engineersfind different customer groups from customer’s basic information.We can get maximizebenefits by making different buying patterns for different customer groups.Clustering analysishas become a powerful research vitality.With the popularization of 3G technology and development of LTE technology,competition of mobile communications companies is becoming intense.The real competitionof the communication market is the competition of business services quality, thecommunication service’s quality is mainly reflected in two aspects.One is the networkstability, the other is customer satisfaction.For network stability and customer satisfaction,thispaper research in network optimization and customer relationship management(CRM).Maincontent include the following:1. Give a summary of development situation in communication industry. Investigated inthe current main data mining clustering algorithm. Analysis distance class,data structure andflow of data mining detailedly.2. According to the concept of customer relationship management ,Research in mobilecommunications industry’s CRM system,applying the data mining technology into the CRMof munication industry.3. Modifying the algorithm into iterative self-organizing data analysis methods based onthe K-means algorithm,using the algorithm cluster a provincial customers bill.According tothe results of data mining ,give the package business proposals.4. Give a introduction on mobile communications network optimization,focusing on thedropped calls of the GSM network problems,give the solutions about the dropped callsproblem.5. Analysis a specifical case of dropped calls by SPSS data mining software.Research thethe relevance between the case data and the historical data.Suggesting the reson factors ofdropped calls,give the treatment options and precautions.