Uses continuous-time stochastic models, mainly based on semi-Markov chains, to derive user state transition patterns, both in rates and in probabilities, in a Web-based information system. Describes search sessions from transaction logs of the University of California's MELVYL library catalog system and discusses sequential dependency. (Author/LRW)

This study developed an analytical approach to detecting groups with homogenous usage patterns in a Web-based information system. Principal component analysis was used for data reduction, cluster analysis for categorizing usage into groups. The methodology was demonstrated and tested using two independent samples of user sessions from the…

Presents a methodology for analyzing patterns of usage of a Web-based library catalog and evaluates the methodology with data collected from the transaction logs of the Web interface version of the University of California's Melvyl Systemwide Library Catalog. Results revealed major differences in usage (number of searches, search time, number of…

Proposes an operational definition of relevance for a Web-based library catalog. Proposes measures that characterize user behavior while searching a Web-based library catalog. Develops a methodology to predict when a user's search will be perceived to be relevant by the user. Tests the methodology with over 900,000 user search sessions with the…