WSF2: A Novel Framework for Filtering Web Spam

Algorithm 2

Dummy filter definition for the WSF2 platform.

(00) web_features SVM check_svm();

() describe SVM Classifies a web page as spam using Support Vector Machine classifier

() score SVM 3

()

() web_features TREE_95 check_tree(0.95, 0.99);

() describe TREE_99 C5.0 between 0.99 and 1.00

() score TREE_99 1.5

()

() web_features TREE_99 check_tree(0.99, 1.00);

() describe TREE_99 C5.0 between 0.99 and 1.00

() score TREE_99 3

()

() web_body HAS_VIAGRA_ON_WEB_BODY eval( "[vV][iI?1!][aA][gG][rR][aA]")

() describe HAS_VIAGRA_ON_WEB_BODY Check if the web page contains references to viagra on body

() score HAS_VIAGRA_ON_WEB_BODY 2

()

() meta HAS_HIGH_SPAM_RATE (SVM & (TREE_95 || TREE_99))

() describe HAS_HIGH_SPAM_RATE Has high probability of being spam

() score HAS_HIGH_SPAM_RATE +

()

() required_score 5