期刊名称：International Journal of Advances in Soft Computing and Its Applications
印刷版ISSN：2074-8523
出版年度：2017
卷号：9
期号：3
页码：23
出版社：International Center for Scientific Research and Studies
摘要：In the telecommunications industry, the competitive intensity forretaining existing customers and avoiding losing valuable customersto competitors has increased dramatically. It is a problem of greatconcern to companies. Customer retention may be boosted bydeploying a prediction model to monitor customer activities. In thispaper, two experiments with the implementation of data processingtechniques using K-Means and Equal-Width Discretization(EWD)combined with Naïve Bayes are performed respectively to conduct acomparison of techniques to identify probable churn activities.Usually, the data generated are of massive size and with highdimensionality.In order to accommodate fast processing, casualheuristics is a preferred deployment. The technique which integrateddifferent algorithm is implemented using Python language under asingle processor environment. By using the correlation betweenattributes, the experimental results show that this can improve themodel in identifying the key factors in churn prediction. The resultshave demonstrated promising overall accuracy.
关键词：Naïve Bayes ; K-Means; data discretization; telecommunications;churn; prediction
