使用彈性詞對匹配與倒傳遞類神經網路之垃圾郵件過濾

Spam Filtering Using Flexible Word Pair Matching and Back-Propagation Neural Network

陳雯玉(Wen-yu Chen)

指導教授 ： 楊燕珠

大同大學/經營設計學院/資訊經營學系所/碩士(2008年)

摘要

本篇提出郵件過濾的方法，使用不同的特徵抽取方式，利用近似樣式匹配選擇出更有意義的詞彙組合，經過特徵值計算，輸入倒傳遞類神經網路，輸出為正常與垃圾郵件兩種類別的預測。以Ling-Spam的郵件作實驗，証實本研究較一些前人的研究有相當好的精確率與召回率。

關鍵字

彈性詞對匹配 ； 倒傳遞類神經網路 ； 垃圾郵件過濾

並列摘要

In this paper we propose a spam filtering method based on different feature extraction, approximate pattern matching, to choose more meaningful lexical combinations. The feature values of each mail are calculated and then feed into Back-Propagation Neural Network to classify as normal or spam mail. Performing experiments on Ling-Spam corpus, the results show that it achieves high precision and recall than several other techniques.

並列關鍵字

Back-Propagation Neural Network ； Spam filtering

參考文獻

16. 許長謨, "從近三年報刊標題看語詞的豐富多變--兼論詞彙學的重要",成大中文學報, vol.11, 2003.

2. H. Drucker, D. Wu, V. Vapnik, “ Support Vector Machines for Spam Categorization”, IEEE Trans Neural Neural Network, 1999.

3. F. Chen , K. Han , G. Chen , “An Approach to Sentence-Selection-Based Text Summarization”, IEEE Region 10 Conference on Computers, Communications, Control and Power Engineering, Oct. 2002.

4. J. Clark , I. Koprinska , J. Poon , “A Neural Network Based Approach to Automated E-mail Classification” , Proceedings of the IEEE/WIC International Conference on Web Intelligence ,2003.

13. L. Zhang , J. Zhu, T. Yao, ”An Evaluation of Statistical Spam Filtering Techniques”, Processing of ACM Transactions on Asian Language Information, December 2004.