Declarative web data extraction and annotation
Carlo Bernardoni, Giacomo Fiumara, Massimo De Marchi, Alessandro Provetti · 2006
Abstract. We propose a software architecture for semantics-based annotation of data extracted from Web sources. Starting from the LiXto suite, which enables semi-automated extraction of XML data from regular documents, we present a solution for attaching background information to individual tags by means of so-called decorations. Decoration is carried out as an inferential activity in the formal context of Answer Set Programming. We discuss a motivating example that will serve as a validation to our approach. 1