| Along with the rapid development of Internet, Intranet, especially WWW, the network, the information resources in the web are becoming more and more abundant. This provides a sound platform for information and resource sharing, has become the largest distributive information warehouse all over the world, and has been a kind of necessary and important access for people to gain the information. However web information has some characteristics, such as vastness, distribution, complexity, lack of consistent manage, which make user unable to completely understand huge and volatile information. Internet users find that it is becoming more and more difficult to search and gain information available,as a large amount of in formation rushes onto the Internet such that a lot of problems, such as "information misnavigation" and "information overloading",emerge. Enginesaer the main tools to help people get information from the WWW. However, the search engines produce many problems: information retrieval based on keywords is low in precision; Documents returned are overabundance and contain redundant information. In this case,the t additional search engines don't cast for the user's desier. Users ex pect to use a new generational search engine,which has intelligence and erpersentation of re sult is concise. Under the huge demand of new information retrieval tools on WWW ,the thesis has an alyzed the development and tendency in the future of the search engine technology, joined to gether with the techniques of Question-Answering,which are used to advance the intelligent ab ility of search engines and provide more humanly human-computer interface. Web-Based Question-Answering system supports natural 1 nguage processing, based on Web information, used concise and precise answers automatically answering user's natural language questions, to help user rapidly find useful informationin thousands upon thousands of WWW information.After analyzing the structure of tranditional QA System and the shortcoming of ordinary process method, the article researchs on QA System technology of Based on Web. And it puts forward QA technology which based frame semantic as semantic base and ontology as knowledge description, and the structure and technology proposal of the prototype system, Book Information QA System Based on Web, which applied into book information domain. The major work of the article shows as follow:(1) Reasearch on the technology of batch collecting, filtering, standard describing for the magnanimity and isomery book information.Develop the Web Book Information Auto-Collecting System.Auto collect, filter and describe book information from multi-website and multi-page, and provide uniform user interface , humanization menu, and function keys, to share different format data, realize function of fetching, saving, searching, browsing, reading and outputting the book information.(2) Reasearch on the question semantic superficial analyse technology in book information domain,including the definition of semantic chunk, the creation of semantic chunk judgement rules,and the definition, analysis, and creation of the question vector.(3) Base on the research of traditional information extraction, research on information extraction technology based on CFN,and apply it into book abstract,to excavate the semantic information in it. The format faced to the traditional information extraction technology always is format or semi-format text, such as html,xml or relational database etc.The reaserach of information extraction technology faced to natural language text is also in primary stage. Because of our Chinese Framenet Semantic Knowledge Base, which is the deep semantic resource, make it possible that to realize information extraction technology based on semantic.(4) Explore on domain ontology building technology, combine the metadata description standard based on Web book information and book abstract conceptual model,used seven-steps as ontology creation method, create book information domain ontology.(5) The Article puts forward the system structure of the Web Book Information QA System,which provides all-around,reliable, high efficiency, intelligentized service for users. For nature language questions printed by users,the system output the book information in accordance with user's request. This proposal prevent the traditional rigid information retrieval technology based on keywords,instead of man-machine conversation and question and ananswer service mode,and provide the flexible, professional, personalize service for users. |
PDF Full Text Request