A Method for Personalized Web Searching with Hierarchical Document Clustering (特集:新世代デ-タベ-ス技術--インタ-ネット・マルチメディア・モ-バイルを中心として)

概要

論文の詳細を見る
In this paper, we propose a new method to customize the information searching on Internet. Existing search tools which evaluate the relevance of documents to the user query give many false positives. One reason is that at query formulation, the information provided by the user is too few to provide good results. In order to complement works on Web agents for autonomously browsing or filtering on behalf of the user, we focus on adaptive agents for providing support at query formulation and retrieval. These adaptive or leaning agents makes possible to add a layer of personalization and collaboration between the users and the existing Web engines. The personalization is realized by learning the user model and using it at query formulation. Information retrieval and clustering mechanisms were utilized to learn the user preferences and to provide support for well-formed personalized query reformulation. We have implemented and experimented our web search method. The results of the experiments show that our approach provide an effective support for customized search and the quality returned results increase.
一般社団法人情報処理学会の論文
1998-04-15