文章基本信息

标题：SMS SPAM DETECTION USING ASSOCIATION RULE MINING BASED ON SMS STRUCTURAL FEATURESNOOR GHAZI M. JAMEEL
本地全文：下载
作者：NOOR GHAZI M. JAMEEL
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2018
卷号：96
期号：12
出版社：Journal of Theoretical and Applied
摘要：The popularity of using mobile phones has led to an increase in sending SMS messages. SMS messages are considered as a rapid way of communication due to its low cost and easy usage. As a result, SMS was target for many types of threats, one of these is spamming. SMS spam is unwanted message sent to many mobile phone users, and cause many problems like annoyance, consuming mobile network bandwidth and other real threats like scam, stealing personal information and installing malware. In this paper, a new system is proposed to detect spam SMSs using Apriori algorithm. This algorithm is used to generate association rules which applied to new SMSs to classify them into spam or legitimate. The system used structural features only instead of textual features or tokens. These features extracted from two publicly available datasets which consists of spam and legitimate SMSs. The rules are generated with different minimum support and minimum confidence values. The generated rules are applied to the test dataset then the rules which achieved higher accuracy are used in the proposed system. The aim of this work is presents a new and fast approach to detect spam SMS using structural features only and to find out if structural features are enough to detect spam SMSs instead of bag of words which depends on preprocessing and consists of many steps like parsing, tokenization, stop word removal and stemming. Good accuracy achieved with 97.65% using rules generated by Apriori association rule mining algorithm with minimum support 0.2 and minimum confidence 0.8 based on SMS structural features only.
关键词：SMS; SMS Spam; Association Rule Mining; Apriori Algorithm