Application of Collocation to Spam Filtering

Author

Zhang, Jing ; Yao, Jianmin ; Dong, Shoubin ; Zhang, Ling

Author_Institution

Sch. of Comput. Sci. & Eng., South China Univ. of Technol., Guangzhou

Volume

2

fYear

2008

fDate

21-22 Dec. 2008

Firstpage

715

Lastpage

718

Abstract

Collocation is the frequent bi-grams of semantic meanings and grammatical functions. Adjacent and long distance collocations are extracted as features for a Bayesian classifier in spam filtering. Compared to the common unigram feature, collocation-based classifier shows improvement in all the evaluation metrics. The influence of mail header information is studied for the classifier, which shows a 10% change in both precision and recall.

Keywords

Bayes methods; feature extraction; information filtering; pattern classification; unsolicited e-mail; Bayesian classifier; adjacent collocations; collocation-based classifier; common unigram feature; evaluation metrics; feature extraction; grammatical functions; long distance collocations; mail header information; semantic meanings; spam filtering; Bayesian methods; Computer science; Computer science education; Educational technology; Geoscience and remote sensing; Information filtering; Information filters; Postal services; Probability; Unsolicited electronic mail; Adjacent collocation; Bayesian classifier; Long-distance collocation; Spam filtering;

fLanguage

English

Publisher

ieee

Conference_Titel

Education Technology and Training, 2008. and 2008 International Workshop on Geoscience and Remote Sensing. ETT and GRS 2008. International Workshop on

Conference_Location

Shanghai

Print_ISBN

978-0-7695-3563-0

Type

conf

DOI

10.1109/ETTandGRS.2008.394

Filename

5070462