Emulate Rule Extraction From Identical Web Sites Based on Rule Ontology
S.C. Punitha, P. Ranjit Jeba Thangaiah, M. Punithavalli · International Journal of Scientific Research · 2012
Web documents are usually semi-structured, and manually processing it is difficult.To overcome the above problem the information contained existing web pages are converted into ontology.We proposed an automatic rule acquisition procedure using ontology, named Rule To Onto, that includes information about the rule components and their struc- tures. We started from the idea that it will be helpful to acquire rules from a site if we have similar rules acquired from other similar sites of the same domain. Rule To Onto is a generalized, condensed, and specifically rearranged version of the existing rules. The rule acquisition procedure consists of the rule component identification step and the rule composition step. We used stemming and seman- tic similarity in the former step and developed the A* algorithm(1) in the latter step. This paper focuses on using Rules Extraction to automatically augment web pages with semantic annotations, and find out whether and how ontology Rules Extractions could be used in the extractionprocess.