Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

15

CEAS
2005
Springer

favoriteEmaildiscussreport

215views Internet Technology» more CEAS 2005»

Naive Bayes Spam Filtering Using Word-Position-Based Attributes

13 years 10 months ago

Naive Bayes Spam Filtering Using Word-Position-Based Attributes

Download www.ceas.cc

This paper explores the use of the naive Bayes classiﬁer as the basis for personalised spam ﬁlters. Several machine learning algorithms, including variants of naive Bayes, have previously been used for this purpose, but the author’s implementation using wordposition-based attribute vectors gave very good results when tested on several publicly available corpora. The eﬀects of various forms of attribute selection—removal of frequent and infrequent words, respectively, and by using mutual information—are investigated. It is also shown how n-grams, with n > 1, may be used to boost classiﬁcation performance. Finally, an eﬃcient weighting scheme for cost-sensitive classiﬁcation is introduced.

Johan Hovold

Real-time Traffic

CEAS 2005 | Naive Bayes | Naive Bayes Classiﬁer | Wordposition-based Attribute Vectors |

claim paper

Related Content

» An evaluation of Naive Bayes variants in contentbased learning for spam filtering

» A Comparison of Event Models for Naive Bayes AntiSpam EMail Filtering

» Spam filters bayes vs chisquared letters vs words

» Spam Email Filtering Using NetworkLevel Properties

» Partitioned logistic regression for spam filtering

» Learning at Low False Positive Rates

» Spam filtering with several novel bayesian classifiers

» Good Word Attacks on Statistical Spam Filters

» Combining Winnow and Orthogonal Sparse Bigrams for Incremental Spam Filtering

Post Info
More Details (n/a)

Added	26 Jun 2010
Updated	26 Jun 2010
Type	Conference
Year	2005
Where	CEAS
Authors	Johan Hovold

Comments (0)