Modeling Web session behavior using cluster analysis: A comparison of three search settings
Dietmar Wolfram, Peiling Wang, Jin Zhang · Proceedings of the American Society for Information Science and Technology · 2007
Abstract Session characteristics taken from large transaction logs of three Web search environments (academic website, public search engine, consumer health information service) are modeled using cluster analysis to determine if different session groups emerge for each environment. The analysis reveals that several distinct clusters of session behaviors emerge, with brief “hit and run” sessions on focused topics, brief sessions on popular topics, and sustained sessions on focused topics with more query modifications. A better understanding of session characteristics can help system designers by developing interfaces or search features that cater to identifiable groups of users based on their search behaviors.