Multi-lingual intelligent information retriever with automated ontology generator

We propose a new information retriever on the Web which: automatically classifies collected documents; and retrieves multi-lingual information (e.g., Japanese and/or English). This is attained by the mechanism of ontology processing. We set up a multi-lingual ontology as an index dictionary. The ontology manages the relations of keywords and the weights of keywords for domain categories to classify the documents. The system has the following features: users can use any keywords expressed in their native language to search for the Web documents, even if they contain multi lingual information; and the documents can be classified into specific domain categories by tuning the keyword weights up.