Conference proceedings article
Logsonomy - Social Information Retrieval with Logdata



Publication Details
Authors:
Stumme, G.; Hotho, A.
Publisher:
ACM
Place:
New York, NY, USA
Publication year:
2008
Pages range:
157-166
Book title:
HT '08: Proceedings of the Nineteenth ACM Conference on Hypertext and Hypermedia

Abstract
Social bookmarking systems constitute an establishedpart of the Web 2.0. In such systemsusers describe bookmarks by keywordscalled tags. The structure behind these socialsystems, called folksonomies, can be viewedas a tripartite hypergraph of user, tag and resourcenodes. This underlying network showsspecific structural properties that explain itsgrowth and the possibility of serendipitousexploration.Today’s search engines represent the gatewayto retrieve information from the World WideWeb. Short queries typically consisting oftwo to three words describe a user’s informationneed. In response to the displayedresults of the search engine, users click onthe links of the result page as they expectthe answer to be of relevance.This clickdata can be represented as a folksonomyin which queries are descriptions ofclicked URLs. The resulting network structure,which we will term logsonomy is verysimilar to the one of folksonomies. In orderto find out about its properties, we analyzethe topological characteristics of th


Keywords
analysis, engine, information, l3s, logsonomy, network, retrieval, search, social, wp5

Last updated on 2019-25-07 at 12:52