With the continuous development of science and technology,the quality of people's life has been improved constantly,especially in the service occupation,better user experience has been pursued by more and more enterprises.In recent years,fraud messages not only has caused widespread concern in the whole society,but also lead many customers to complained.Three operators,as the industry leader,it is exactly the task of top priority to solve the problem and it is an imminent to create a better user experience.In this article,we will use the method of machine learning to classify the telecom operators of 800000 text message,then compare to the method of sentiment analysis algorithm.According to the returned results,the na?ve Bayesian algorithm which is used in the text classification is more accurate and reasonable.At last,the paper is intended to give some suggestions on SMS fraud for the operators from the classification and prediction effect. |