Generating New Features Using Genetic Programming to Detect Link Spam
Created by W.Langdon from
gp-bibliography.bib Revision:1.8098
- @InProceedings{Li:2011:ICICTA,
-
author = "Shengen Li and Xiaofei Niu and Peiqi Li and Lin Wang",
-
title = "Generating New Features Using Genetic Programming to
Detect Link Spam",
-
booktitle = "2011 International Conference on Intelligent
Computation Technology and Automation (ICICTA)",
-
year = "2011",
-
month = mar,
-
volume = "1",
-
pages = "135--138",
-
abstract = "Link spam techniques can enable some pages to achieve
higher-than-deserved rankings in the results of a
search engine. They negatively affect the quality of
search results. Classification methods can detect link
spam. For classification problem, features play an
important role. This paper proposes to derive new
features using genetic programming from existing
link-based features and use the new features as the
inputs to SVM and GP classifiers for the identification
of link spam. Experiments on WEBSPAM-UK2006 show that
the classification results of the classifiers that use
10 newly generated features are much better than those
of the classifiers that use original 41 link-based
features and equivalent to those of the classifiers
that use 138 transformed link-based features. The newly
generated features can improve the link spam
classification performance.",
-
keywords = "genetic algorithms, genetic programming, GP
classifier, SVM, WEBSPAM-UK2006, classification method,
link spam detection, link-based feature generation,
search engine, search result quality, Internet, feature
extraction, information retrieval, pattern
classification, search engines, support vector
machines",
-
DOI = "doi:10.1109/ICICTA.2011.41",
-
notes = "Also known as \cite{5750574}",
- }
Genetic Programming entries for
Shengen Li
Xiaofei Niu
Peiqi Li
Lin Wang
Citations