Modeling Implicit Knowledge from Microformatted Websites.

Abstract—Microformats are a technique to incorporatesemantics into web documents by means of standard XHTMLtags enriched with particular attributes. They are a set ofsimple and open metadata that describe semantic units ofinformation which are called classes.Despite of its practicality, microformats lack of a formalmodel for representing relations between semantic units ofinformation, and hence, they do not describe knowledge.This fact contrasts with semantic technologies that alreadyrepresent knowledge, like for instance, RDF.In order to face this problem, in this work we introducea formal model able to represent the classes provided bymicroformats, and their relations. In essence, we modelthe semantic information, in a graph like structure, namelysemantic network, where edges are labeled with predicates.We show how semantic networks allow us to model microfor-mats in a very convenient way for reasoning and knowledgeextraction.Keywords: Semantic Web, Microformats, Information retrieval.