Back to Results
First PageMeta Content
Science / Natural language processing / Machine learning / Statistical classification / Email / Email filtering / Naive Bayes classifier / Language model / N-gram / Statistics / Spam filtering / Computing


An Adaptive, Semi-Structured Language Model Approach to Spam Filtering on a New Corpus Ben Medlock Cambridge University Computer Laboratory William Gates Building JJ Thomson Avenue
Add to Reading List

Document Date: 2007-12-07 05:59:37

Open Document

File Size: 220,73 KB

Share Result on Facebook

City

Ann Arbor / New York / Cambridge / Mountain View / /

Company

A. Genkin D. D. / Neural Networks / ILM / MIT Press / Unigram ILM Bigram GEN / Enron / ACM Press / Springer-Verlag New York Inc. / SPAM / TP T / Advances / MessageLabs / /

Country

United States / /

/

Event

Product Issues / Product Recall / /

IndustryTerm

hill-climbing algorithm / approximation algorithm / data mining / internet urls / /

Movie

Test / /

Organization

MIT / idf / University of Cambridge Millennium Scholarship / New Corpus Ben Medlock Cambridge University Computer Laboratory / Association for Computational Linguistics / /

Person

Mark Gales / C. Burges / Thomas Hain / G. Paliouras / V / Ted Briscoe / /

Position

editor / author / Semi-Structured Language Model Approach / representative / Bayesian network model for semi-structured document classification / /

Product

Unigram ILM Bigram MNB SVM BLR ILM Unigram ILM Bigram GEN / threshold / values / improvement / 0.6 0.7 0.7 0.8 0.8 0.9 0.9 1.0 1.0 Adaptive Component Weight Figure / 0.91 0.9 1 0.99 0.98 0.97 0.96 0.95 GEN / Accuracy / /

ProgrammingLanguage

XML / /

ProvinceOrState

New Brunswick / New York / Michigan / Massachusetts / /

PublishedMedium

Computational Linguistics / Machine Learning / /

Technology

XML / speech recognition / LM technology / perl / Machine Learning / AdaBoost algorithm / hill-climbing algorithm / approximation algorithm / data mining / /

URL

http /

SocialTag