An adaptive short list for documents on the World Wide Web

University of Maribor School of Psychology University of MariborSmetanova 17 Georgia Institute of Technology Smetanova 17S1-2000 Maribor, Slovenia Atlanta, GA 30332 USA S1-2000 Maribor, Slovenia+386 62221 112 +1 4048948344 +386 62221 112debevc@uni-mb. si bmeyer@psy.tfe. gatech. edu rajko. svecko@tni-mb. siABSTRACTSince the World Wide Web (WWW) is so popular andgrowing so quickly, users have almost an intinite numberof sites to choose from. Bookmark features in web browsersallow users to easily record the sites that they would like tobe able to view again, without having to repeatedly searchthrough the WWW. However, bookmark lists for activeweb users can grow very long very quickly. Since user-maintained bookmark lists can easily grow long andsomewhat harder to use, it is usefil to have anautomatically maintained shorter list of useful sites.This paper describes an Adaptive Short List of commonlyused sites. This feature, when integrated into web browsingsoftware. would enable users to check the most probablesites quickly. without having to search through everybookmark they’ve ever created. We also present a decisionalgorithm for selecting sites to include in this list. Thegoal of this system is to determine the most appropriatesites to include in the Adaptive Short List. based on usagedata which the system collects and analyzes while the userworks.KeywordsUser interface, adaptive user interface, World Wide Web,intelligent system. user rnodellingINTRODUCTIONAs more and more information becomes available on theWorld Wide Web and as users discover more of thisinformation, the number of web sites to which they theywant ready access also grows. Web browsing software suchas Netscape [4] supports this need by providing a“bookmark feature. allowing users to return directly to