Social networks and web minorities

• Published in: Cognitive systems research Vol. 4; no. 4; pp. 355 - 364
• Main Authors: Gori, Marco, Numerico, Teresa
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2003
• Subjects: Crawling the Web; Intelligent agents; Learning from examples; Mining the Web; PageRank; Web minorities; PageRank; Crawling the Web; Mining the Web; Web minorities; Intelligent agents; Learning from examples
• ISSN: 1389-0417; 1389-0417
• DOI: 10.1016/S1389-0417(03)00016-0

The conception of appropriate models of information retrieval is a crucial step for the actual growth and development of the web. In addition to traditional information retrieval techniques, nowadays search engines have adopted methods for taking into account the social network deriving from the pattern of interconnections. Using different arguments, it has been shown that this is very effective in practice. In this paper we subscribe that point of view, but state also that social networks based ranking schemes operating on the whole web, like Google’s pagerank, prevent small communities to become visible on the Web, regardless of the quality of the information. Hence, we strongly advocate the development of new cognitive models of the whole process of information retrieval which are capable of facing this problem. We give a picture of distributed architectures which incorporates topic-based intelligent agents and personal agents for capturing the user’s profile.