The aim of this study is to investigate relationship between past users’ behavior (described by access patterns) and future one. The two main ideas are first to explore the possible users’ characterization that can be extracted from access pattern. This allows to measure and to have a better understanding of users’ behavior. This knowledge allows us to build new services as building interest communities based on a comparative approach and clustering. The second idea is to see if these characterizations can be useful to forecast future access. This could be useful to prefetch web data in proxy-cache. We show that there are some partial mathematical models binding the users’ behavior to the repetition of queries. © Springer-Verlag Berlin Heidelberg 2002.
CITATION STYLE
Durand, N., & Lancieri, L. (2002). Study of the regularity of the users’ internet accesses. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2412, 173–178. https://doi.org/10.1007/3-540-45675-9_29
Mendeley helps you to discover research relevant for your work.