應用分類迴歸樹區別通訊上未經請求之電子郵件

10.29614/DRMM.201110.0002

葉振山；林器弘

分類廻歸樹 ； 未經請求之郵件 ； 垃圾郵件 ； 電子郵件病毒 ； 受測者特性操作曲線

資訊安全通訊

17卷4期（2011 / 10 / 01）

23 - 42

繁體中文

網際網路的基礎設施的普及使得電子郵件使用已經成為每日生活的一部分，但透過網際網路為基礎的電子郵件系統，常被未經請求的信件淹没，並且這些郵件中含有惡意電子郵件散佈病毒很難加以防範，儘管偵測垃圾郵件和惡意碼的演算法和架構研究也不斷的提出，未經請求的信件仍然能夠避免被偵測。這些未能被偵測或惡意的信件造成使用者耗費大量時間和成本去刪除和管理電子郵件，並且危及資訊安全與浪費頻寬資源。本研究利用公開之樣本，萃取郵件表頭、超連結、與收發郵件行為等17種特徵以分類廻歸樹之分類方法做識別未經請求電子郵件並和其他四種方法比較，結果我們得出18條規則並且決策樹分類優於其他四種方法，並且正確率（預測率）為93%，假陽性率與假陰性率分別為0.02與0.04。

1. CSMining Group. Spam email datasets. Available http://csming.org/index.php/spam-email-datasets-.html
2. Abe, Tsuyoshi,Miyake, Jun,Kawashima, Masahisa,Takahashi, Katsumi(2005).Spam Filtering with Cryptographic Ad-hoc Email Address.Proceedings of the Symposium on Applications and the Internet Workshops
3. Androutsopoulos, Ion,Koutsias, John,Chandrinos, Konstantinos V.,Spyropoulos, Constantine D.(2000).An experimental comparison of naive Bayesian and keyword-based anti-spam filtering with personal e-mail messages.Proceedings of the 23rd annual international ACM SIGIR conference on Research and development in information retrieval,Athens, Greece:
4. Chirita, Paul-Alexandru,Diederich, Jorg,Nejdl, Wolfgang(2009).MailRank: Using Ranking for Spam Detection.ACM Transactions on Knowledge Discovery from Data,3(3),373-380.
5. Crawford, E.,Kay, J.,McCreath, E.(2001).Automatic Induction of Rules for e-mail Classification.ADCS2001 Proceedings of the Sixth Australasian Document Computing Symposium,Coffs Harbour, NSW Australia:
6. Crocker, Dave.Challenges in Anti-Spam Efforts.The Internet Protocol Journal,8(4),1-31.
7. Duan, Zhenhai,Dong, Yingfei,Gopalan, Kartik(2007).DMTP: Controlling spam through message delivery differentiation.ComputerNetworks,51,2616-2630.
8. El-Alfy, El-Sayed M.,Abdel-Aal, Radwan E.(2011).Using GMDH-based networks for improved spam detection and email feature analysis.Applied Soft Computing,11,477-488.
9. Glymin, Mawuena,Ziarko, Wojciech(2007).Rough Set Approach to Spam Filter Learning.Leciture Notes in Artificial Intelligent,4585,350-359.
10. Lai, Chih-Chin,Tsai, Ming-Chi(2004).An Empirical Performance Comparison of Machine Learning Methods for Spam E-mail Categorization.Proceedings of the Fourth International Conference on Hybrid Intelligent Systems 04'
11. Lai, Gu-Hsin,Chen, Ghia-Mei,Laih, Chi-Sung,Chen, Tsuhan(2009).A Collaborative anti-spam system.Expert Systems with Applications,36,6645-6653.
12. Park, Insu,Sharman, R.,Rao, H. R.,Upadhyaya, S.(2007).Short Term and Total Life Impact analysis of email worms in computer systems.Decision Support Systems,43,827-841.
13. Sahami, Mehran,Dumais, Susan,Heckerman, David,Horvitz, Eric(1998).A Bayesian approach to filtering junk e-mail.AAAI-98 Workshop on Learning for Text Categorization
14. Sasaki, M.,Shinnnou, H.(2005).Spam detection using text clustering.Cyberworlds, 2005. International Conference
15. Shih, Dong-Her,Chiang, Hsiu-Sen,Yen, C. David(2005).Classification methods in the detection of new malicious emails.Information Sciences,172,241-261.
16. Stolfo, Salvatore J.,Hershkop, Shlomo,Wang, Ke,Nimeskern, Olivier,Hu, Chia-Wei(2003).Behavior Profiling of Email.Lecture Notes in Computer Science,2665,87-90.
17. Wu, Chih-Hung(2009).Behavior-based spam detection using a hybrid method of rule-based technologies and neural networks.Expert Systems with Applications,36,4321-4330.
18. Yu, Bo,Xu, Zong-ben(2008).A comparative study for content-based dynamic spam classification using four machine learning algorithms.Knowledge-Based Systems,21,355-362.
19. Yu, Bo,Zhu, Dong-hua(2009).Combining neural networks and semantic feature space for email classification.Knowledge-Based Systems,22,376-381.
20. Zhang, L. E.,Zhu, Jingbo,Yao, Tianshun(2004).An Evaluation of Statistical Spam Filtering Technoiques.ACM Transactions on Asian Language Information Processing,3(4),243-269.
21. Zhoua, Jianying,Chin, Wee-Yung,Roman, Rodrigo,Lopez, Javier(2007).An effect multi-layered defense framework against spam.Information Security Technical Report,12,179-185.