Phone Spam Filtering System

Along with the mobile communication technology rapid development and the handset popular rate rapid enhancement, the handset short message more and more receives the multitudinous public figure's favour. The short message industry takes up honest work by its short, rapid, simple, the price is inexpensive and so on many merits to become people's one kind of important correspondence and the exchange way day by day. Motion short message service in rapid development process, on the one hand has brought each kind of convenience; On the other hand as a result of each kind of trash short message massive appearances, also has brought in the information security question, has affected people's normal life. Thus extremely has the necessity to filter out trash short message.Through to the short message service structure analysis, the short message transmission way has two kinds, one kind is the SP- handset, moreover one kind is the handset - handset, regardless of which one kind all is carries on the memory through the short message service center to retransmit, again arrives the user handset. Therefore may carry on from two aspects to trash short message filters, one kind is the short news service center filters the way, another kind is the handset terminal filters the way. Regardless of is filters the way from the short news service center or the handset terminal to filter the way, all has some insufficiencies.In order to the better realization trash short message filters, this system used chooses the handset terminal to filter the way, unified the method through the blacklist technology and the text classification technology, use Naive Bayes classifies the method to come to carry on the classified recognition to the short message content, thus realized to trash short message filtering.The full text has given trash short note filtration system design proposal, has given the detailed elaboration and the design to some key technologies,finally has carried on the result analysis and the appraisal to the Naive Bayes classified model.