End user searching on the Internet: An analysis of term pair topics submitted to the Excite search engine

Queries submitted to the Excite search engine were analyzed for subject con tent based on the cooccurrence of terms within multiterm queries. More than 1000 of the most frequently cooccurring term pairs were categorized into o ne or more of 30 developed subject areas. Subject area frequencies and thei r cooccurrences with one another were tallied and analyzed using hierarchic al cluster analysis and multidimensional scaling. The cluster analyses reve aled several anticipated and a few unanticipated groupings of subjects, res ulting in several well-defined high-level clusters of broad subject areas. Multidimensional scaling of subject cooccurrences revealed similar relation ships among the different subject categories. Applications that arise from a better understanding of the topics users search and their relationships a re discussed.