An Interactive Classification of Web Documents by Self-Organizing Maps and Search Engines

In this paper, we propose an effective classification view mechanism for hypertext data such as web documents based on Kohonen’s Self-Organizing Map (SOM) and search engines. Web documents collected by search engines are automatically classified by SOM and the obtained SOMs are incrementally modified according to the interaction between users and SOMs. At present, various search engines are designed to retrieve web documents. When we use search engines to retrieve web documents, we get a lot of answers as ever before, so we have a lot of labors to examine each web document. Therefore, in order to make up for search engines, we need a function to classify web document corresponding to the user’s point of view and their purposes. Furthermore, we cannot retrieve pertinent web documents by conventional search engines when a specific topic is described by more than one web document. To solve these problems, we exploited a contentbased clustering system for web documents. In this system, web documents are automatically clustered by their feature vectors produced from web documents or minimal subgraphs consisting of multiple web documents, and their overview maps are dynamically generated by SOM. Furthermore, we propose a method by which an obtained SOM is modified by user’s interaction such as feedback operations. It is important to reflect the aim of classification and the purpose of retrieval to this system. In our research, we intend to solve these problems by providing a view mechanism in which the Basic Units for retrieval and clustering of Web Documents (BUWDs) are changeable by users and relevance feedback operations enable the generation of an overview map which reflects user needs.