Research Of Intelligent Answering System Based On Chinese Word Segmentation Amd Ontology

At present,network teaching has gradually become an effective way for people to learn.As the limitations of the network teaching itself,question-answering system become some useful supplement of it.The question answering system is also becoming a hot issue in the field of Natural Language Processing.With the quickly development of information and artificial intelligence technology,the different realization of online question answering system mode emerge.At the present,the question-answering system in China is still of limitation,mainly reflected in two aspects:not enough intelligence and lack of standardization of knowledge base.Not enough intelligence is mainly reflected in the use of the logic combination of keywords when asking questions and the lack of natural language support.Lack of standardization of the knowledge base makes it difficult to be shared and reused.Ontology possesses a favorable concept hierarchical structure and sustain logical reasoning.The content of the document can be marked by the relationship between the concepts,so the ontology can express semantics,which makes it possible to retrieve education resources based on content.An increasing number of scholars are trying to apply ontology to information retrieval and intelligent question answering system.Three aspects of the work are finished in this paper.First,the present research situation of intelligent question answering system at home and abroad is provided,and the existing problems in the research of intelligent question answering system is pointed out.The research status on the related technology,such as,the ontology technology and Chinese segmentation technology,are reviewed,mainly concentrated in ontology description languages,methodology and means of constructing of ontology,the Chinese segmentation algorithms and their issues.Secondary,combined with the construction experience and methods of successful ontology engineering at present,according to the Seven Steps Method,developed by the Stanford University,with building tools of ontology,Protégé,we construct a small course Ontology--"course knowledge ontology of computer network".Finally,be aimed at the current problems of intelligent question answering system,an intelligent question answering system model based on ontology and Chinese word segmentation technique is put forward.In this model,the user can ask question with Chinese natural language,then the system will automatically segment the words for the question.After expanding the keywords,the SPARQL query will be constructed,and used to find the answer of the questions in the ontology library.If the answer can't be found in the ontology library,then the similar documents from document library will be found and presented to the user by the similarity degree of the documents.In this model,a multi-level Chinese segmentation algorithm is adopted,i.e,techniche terms will be segmented first based on erminological dictionary,and then signal words will be segmented based on signed word s dictionary,in the last the plain words will be segmented.