垃圾郵件過濾器之研究—以貝氏推論為基礎

隨著網際網路快速的蓬勃發展，電子郵件便成為現代人交流通訊不可或缺的工具，但也因此也成為不請自來的行銷郵件、病毒、惡意程式…等等的傳播媒介，對個人的電腦，甚至企業的電腦造成很大的威脅，而且有越演越烈的趨勢，所以如何處理電子郵件的安全問題是現今網際網路問題的重要課題之一。本研究先就垃圾郵件的危害作分析，並提出現行垃圾郵件過濾的方法，最後採用貝氏推論中的天真貝氏分類法和貝氏網分類法來建構過濾器，希望可建構出比現在已知的貝氏分類法為基礎更好的過濾器。

關鍵字

電子郵件；垃圾郵件；簡單貝氏；貝氏網；過濾器

並列摘要

With the fast development of Internet, e-mail has been becoming the most and fastest economical modern way for communication. But it begins to be the carrier of spam、virus or the malicious program, which seriously threaten the safety of the private and business computer system. So how to solve its security problem properly has become one of important topic. First, it essay discusses the situation and the harm of spam and present spam-filter of spam. Finally, the uses Native Bayes and Bayes Net to establish spam filter and hope it can be better than the current Bayes used.

並列關鍵字

e-mail ； spam ； Na&iuml ； ve Bayes ； Bayes Net ； filter

參考文獻

[1] Akaho Shotaro,(2002)” Conditionallyindependent component analysis for supervised feature extraction”, Neurocomputing vol.49 pp.139 – 150

[4] Chib Siddhartha, Federico Nardari, Neil Shephard,” Markov chain Monte Carlo methods for stochastic volatility models(2002)”, Journal of Econometrics vol.108 pp.281 – 316

[5] Dorp J. Rene Van, Thomas A. Mazzuchi,(2004)” A general Bayes exponential inference model for accelerated life testing”, Journal of Statistical Planning andInference vol.119 pp.55 – 74

[6] Karunamuni Rohana J., Shunpu Zhang,(2003)” Optimal linear Bayes and empirical Bayes estimation and prediction of the &nite population mean”, Journal of Statistical Planning and Inference vol.113 pp.505 – 525

[7] Madi Mohamed T., Thomas Leonard, Kam-Wah Tsui,(2000)”Bayes inference for treatment effects with uncertain order constraints”, Statistics & Probability Letters vol.49 pp.277 - 283

國際替代計量

垃圾郵件過濾器之研究—以貝氏推論為基礎

未授權

主題瀏覽