| 摘要 |
he system disclosed herein is composed of 4 subsystems: the Thematic Crawler (TC), the Information Miner (IM), the Document Organizer (DO), the Database (DB) for the storage of the information, and the knowledge database (WordNet-WN) all installed on the same or different computers interconnected on a local network. The system is installed on an Internet connected computer and allows the automatic collection of thematic documents from the Internet by using ontologies and the hyperlinks interconnecting the Internet documents. At the entry of a thematic ontology O1, the information-collecting subsystem exports a collection of internetic documents. In the sequel, the export and information enrichment subsystem exports the most significant words of each document and sets them in correspondence to the ontology meanings. The document-organizing subsystem creates automatically batch documents, which are classified in non-predefined categories with respect to the similarity of the meanings of same documents. The information produced by the two subsystems IN & DO is stored on the database (DB). The query-answering subsystem (Query Manager-QM) receives at its entry queries (Q3) and answers them by supplying documents with meanings similarto those of the query. The documents are presented in a classified batch form (R3). |
