Font Size: a A A

Android-based Spam Messages Processing System

Posted on:2013-11-11Degree:MasterType:Thesis
Country:ChinaCandidate:C ZhangFull Text:PDF
GTID:2248330377455292Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
The short message industry takes up honest work by its short, rapid, simple, the price isinexpensive and so on many merits to become people’s one kind of important correspondence andthe exchange way day by day. Motion short message service in rapid development process, on theone hand has brought each kind of convenience; On the other hand as a result of each kind of trashshort message massive appearances, also has brought in the information security question, hasaffected people’s normal life. The so-called hot spots short message is that the people have beenmost concerned about a stage in a certain category of mobile phone short message, and people aremost concerned about now is continue to be received of the messages. This paper on the currentspam become a social hot spot of the status quo, focusing on the classification of the trash messages,as well as incremental feedback solutions for users of SMS spam different criteria for judging.Firstly, this paper introduces the development status quo of trashy short message and nowAnti-trashy message technology, as well as the basic concepts and principles of short messagefiltering. Then introduce the principle of Bayesian classification mainly. Analyze the limitationexisted in short message filtering using the traditional Bayesian algorithm.(Legitimate shortmessage is misjudged which can bring user greater losses.) On this basis, we adopt the improvedNaive Bayes and black-and-white list to filter messages. According to personalized feedbackautomatic incremental learning classifier further enhance accuracy. The experimental results onChinese message corpus show that this algorithm correctly classifies short messages at the sametime, legitimate messages can also reduce false alarms. We obtain good performance whenclassifying and filtering short messages.This article contains mainly the following contents:1) Based on the built-in short message corpus, Introduce a improved Bayesian classificationmodel that reduced rate of normal messages misjudgment, NB is used to carry outexperiment on Chinese merge corpus, Analyze and compare the experimental results andperformance.2) Design of a Short Message Filtering System base on Android OS, and according to theindividual requirements of users automatically generate dynamic feedback filteringrules,SMS filtering system to meet the users in the dynamic change of classification criteria requirements;3) Under the real conditions that there is no an open and standardized Chinese messagecorpus, a true and standardized one that is able to adapt to experiment is established;4) Summarize the status quo of trashy message filtering investigating, including thedefinition of trashy message and generating, as well as the filtering technology used often.
Keywords/Search Tags:Android, Naive bayes, Text classification, Trash Short Message
PDF Full Text Request
Related items