Research On Sms Filtering Algorithm Based On User Features

With the popularity of smart phones and the rise of mobile e-commerce,SMS(Short Message Service)function is no longer limited to personal communication,but is widely used in accessing dynamic password,getting express information,advertising and so on.With the rapid development of mobile SMS business,the demand for SMS filtering has undergone profound changes.The existing SMS filtering technology is aimed at the traditional sense of the normal text messages and spam messages.And the SMS filter designed without fully considering the users individual needs.In the popularity of personalized service today,SMS filtering technology need to keep up with the pace of the times.The purpose of this thesis is to study the user characteristic modeling technology and classification algorithm,combined with the specific application scenario of SMS filtering,and put forward the whole solution of SMS filtering for intelligent terminal equipment,and finally build the filtering system.This thesis mainly studied the text categorization algorithm,and proposed a user feature model which is based on the apps category messages collected from the user's intelligent terminal handset,put forward another user feature model which is based on the user's operation upon his or her existing SMS messages and a SMS filtering algorithm based on these user characteristic models.The works done in this thesis are as follows:1)Established the user characteristic model based on the app category according to the app install,use,uninstall and share informations collected from the user's smartphone;2)Established the user characteristic model based on the user's operation upon existing SMS messages according to the existing short messages in the user's smart phone and the user's operation information of these short messages;3)Designed the SMS filtering algorithm according to the user characteristics model based on the App category and the user characteristic model based on the user's operation upon existing SMS messages,and realized it.4)Designed comparison experiment for testing and analysis on the SMS filtering algorithm,and verified the SMS filtering algorithm's feasibility and effectiveness.This thesis focuses on the in-depth study of individualized SMS filtering model,and achieved the corresponding model based on the Android platform,which has a certain practical value and reference significance to improve the overall performance of SMS filtering system.