23 February 2005 User profiling in WWW network
Author Affiliations +
Abstract
In this paper we present application of several theoretical tools to a problem of generation of WWW user pofile. First, we present the idea of user profiling and the problem of classification using decision rules. The user profile consists of rules concerning time of day, amount of pages that were browsed and presumable time spent on browsing pages in one WWW host. This profile can be used for planning of advertisement on WWW sites. In the next section, we shortly describe selected tools for generation of decision rules that were used to create user profile. We have chosen two tools -- CART and C4.5 that are examples of decision tree algorithms. We also present the process of data mining. The following section of this paper presents the discussion of results.
© (2005) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michal Kekez, Adam Gluszek, Filip Rudzinski, "User profiling in WWW network", Proc. SPIE 5775, Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments III, (23 February 2005); doi: 10.1117/12.610758; https://doi.org/10.1117/12.610758
PROCEEDINGS
4 PAGES


SHARE
Back to Top