Using latent semantic indexing for information filtering
COCS '90 Proceedings of the ACM SIGOIS and IEEE CS TC-OA conference on Office information systems
Large-scale information retrieval with latent semantic indexing
Information Sciences: an International Journal
In search of reliable usage data on the WWW
Selected papers from the sixth international conference on World Wide Web
TalkMine and the adaptive recommendation project
Proceedings of the fourth ACM conference on Digital libraries
Determining the publication impact of a digital library
Journal of the American Society for Information Science - Special topic issue on digital libraries: part 2
ECDL '00 Proceedings of the 4th European Conference on Research and Advanced Technology for Digital Libraries
Group User Models for Personalized Hyperlink Recommendations
AH '00 Proceedings of the International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems
Constructing Web User Profiles: A non-invasive Learning Approach
WEBKDD '99 Revised Papers from the International Workshop on Web Usage Analysis and User Profiling
Distributed, real-time computation of community preferences
Proceedings of the sixteenth ACM conference on Hypertext and hypermedia
Hi-index | 0.00 |
We propose a methodology to evaluate the impact of a Digital Library's (DL) collection and the characteristics of its user community by an analysis of user retrieval patterns. Patterns of journal and document co-retrievals are reconstructed from DL server logs and used to generate proximity data for journals and documents, resulting in a weighted relation defined over the DL document collection represented by a network of document and journals. A measure of discrepancy between user-defined measures of document impact and the Journal Citation Record (JCR) Impact Factor (IF) published by the Institute for Scientific Information (ISI) is used to analyze characteristics of the DL user community. A preliminary analysis of the Los Alamos National Laboratory (LANL) Research Library (RL) server logs registered in 2001 demonstrates the potential of this approach.