Phrase Grammar-based Automatic Conceptual Tagging System

摘要

The advent of XML framework has brought into focus the notion of semantic tagging of Web documents. At present, many of the web pages are in the HTML format, and the next generation of the Web is likely to be dominated by XML. In this context, a mechanism to automatically detect and tag domain-specific concepts in a classified set of documents will reduce the manual effort needed to tag the free text documents. In this paper, we describe a mechanism to accomplish this using a phrase grammar-based approach. The referral link-based ranking method of web pages is critically examined. The proposed method is projected as a semantic alternative to this ranking scheme.