Abstract

We present an approach for detecting link spam common in blog comments by comparing the language models used in the blog post, the comment, and pages linked by the comments. In contrast to other link spam filtering approaches, our method requires no training, no hard-coded rule sets, and no knowledge of complete-web connectivity. Preliminary experiments with identification of typical blog spam show promising results.

Citations

684	A language modeling approach to information retrieval - Ponte, Croft - 1998
309	E: A bayesian approach to filtering junk e-mail. AAAAI’98 Workshop on Learning for Text Categorisation - Sahami, Dumais, et al. - 1998
97	Relevance based language models - Lavrenko, Croft - 2001
86	Challenges in web search engines - Henzinger, Motwani, et al. - 2002
80	Telling humans and computers apart automatically - Ahn, Blum, et al. - 2004
74	An experimental comparison of naive Bayesian and keyword-based anti-spam filtering with personal e-mail messages - Androutsopoulos, Koutsias, et al. - 2000
67	Damn Spam, and Statistics, Using statistical analysis to locate spam web pages - Spam
58	Retrieval and novelty detection at the sentence level - Allan, Wade, et al. - 2003
58	Recognizing nepotistic links on the web - Davison - 2000
45	The connectivity sonar: Detecting site functionality by structural patterns - Amitay, Carmel, et al. - 2003
19	Web Page Ranking Using Link Attributes - Baeza-Yates, Davis - 2004
14	Evaluating cost-sensitive unsolicited bulk email categorization - Hidalgo - 2002
1	nofollow: fight spam, not blogs - No