Issue No. 01 - January/February (2003 vol. 18)
Roberto Basili , University of Rome Tor Vergata
Alessandro Moschitti , University of Rome Tor Vergata
Maria Teresa Pazienza , University of Rome Tor Vergata
Fabio Mossimo Zanzotto , University of Rome Tor Vergata
<p>The need for quick and personalized access to distributed information is currently amplified by the growing capabilities of fast and global computer and telecommunications networks. As this empowers the penetration of more sophisticated information management practice, it also increases the complexity of the required information technologies. Multimedia applications for Web-based information access provide high-level interfaces and facilities but still rely on poorly harmonized multimodal front ends. Search engines, for example, are still tied to traditional keyword-based methods and depend loosely on Web content. The authors are convinced that more emphasis on the linguistic content is a viable path to intelligent Web services. Textual information can provide a logical approach to Web authoring where hypertext links can be inferred from conceptual information extracted from the texts. They have pursued these ideas in Namic, a multilingual text classification and hyperlinking system based an a knowledge-based approach to information extraction. This article describes the Namic architecture and its application scenario.</p>
natural language processing, information extraction, knowledge management, Web publishing
F. M. Zanzotto, M. T. Pazienza, R. Basili and A. Moschitti, "Personalizing Web Publishing via Information Extraction," in IEEE Intelligent Systems, vol. 18, no. , pp. 62-70, 2003.