The Key Technologies For Construction Of Uyghur Domain Ontology And Applications

With the deepening and broader development of knowledge discovery related technology research,ontology methods begin to expose their potential defects and limitations.Because the Internet has spread all over the world,how to obtain knowledge from the information resources in the fastest and most accurate way has become an important topic that has to be faced nowadays.Under the background of the continuous development of retrieval technology,the retrieval efficiency has been significantly improved,but it still can not reach people's expectations from the retrieval results.Therefore,the normative organization of knowledge is also an important object of people's attention.The birth of the ontology has made the knowledge organization have good conditions.However,there is almost no ontology-related research report and available ontology resources about Uyghur language.The ontological research of Uyghur language has just started.Moreover,the existence of various ways and domain differentiation in the construction ontology has made the ontology sharing and reuse limited.To build an ontology construction specification for this purpose is an important prerequisite for the smooth construction and large-scale development of the ontology.In order to ensure that the ontology of knowledge organization can exert its greatest advantage,it can create favorable conditions for knowledge analysis,knowledge retrieval and knowledge storage.Aiming at this situation,the ontology construction abstract method is used as the research guidance of this paper,and specifically divides the Uyghur domain ontology construction work.And it includes the initial artificial construction of the Uyghur domain ontology,the semi-automatic construction of the general ontology knowledge base UWN and other applications.On this basis,the domain ontology is automatically constructed through concept acquisition and relationship acquisition,and finally applied to the Uyghur automatic summarization.The specific work of this paper includes the following contents:1? In-depth analysis of ontology definitions,related theories and methods.Firstly,the definition of ontology is explained and studied,and the characteristics of ontology in knowledge sharing and description are analyzed.The basic elements,classification,construction methods,description language and construction tools of ontology are expounded.2?The manual construction process of Uyghur domain ontology is introduced in detail.This paper selects Protege4.3 tool and OWL ontology description language,and uses the improved seven-step method,combined with the language characteristics of Uyghur language,to realize the Uyghur language ontology library in the field of that mathematics and information science.On this basis,the Jena toolkit is used to semi-automatically constructed the Chinese-Uyghur bilingual ontology in the field of University Management and to implemented the query of SPARQL language.Then,this paper uses the cross-lingual reuse on domain ontology construction method to extract the triples of the collected English domain ontology collection.Then use cross-lingual technology to match the English and Uyghur concepts and relationships,and then implement the Uyghur standard domain ontology through the Jena tool.The experimental results verify the grammatical accuracy of the constructed domain ontology,and at the same time,the Jena open source project is used to build the domain ontology construction platform,which lays a foundation for future research work.3?Due to the need of research,there is a lack of a thesaurus or a structured knowledge base such as WordNet and HowNet in Uygur language.Therefore,the Uyghur WordNet(Abbreviated as UWN)general ontology knowledge base is developed.It has a matching rate of 90% with the English WordNet concept.The part of speech also considers nouns,verbs,adjectives and adverbs.The relations only be considered the hypernyms and hyponyms relations,synonyms and antonyms.Based on this work,UWN is reused to expand and enrich the Uyghur tourism domain ontology.4?The automatic construction of Uyghur domain ontology is realized on the basis of the preliminary work.The concept extraction method of multi-feature fusion and the relations extraction based on hybrid hierarchical clustering method are used to realize the automatic construction of Uyghur tourism ontology.Due to the limitations of the text content selected in this paper and the difficulty of merging nodes by hierarchical clustering,although the hierarchical clustering results are not ideal,but they basically conform to the ontology hierarchy.The experimental results show that the method is feasible,and the factors such as increasing the vector dimension and the text content can further improve the accuracy of the domain concepts and relations.5?The constructed domain ontology is integrated into the automatic summary extraction.In this paper,the ontology technology is introduced into the Uyghur automatic abstract extraction,and a high-quality Uyghur automatic summary based on the ontology is obtained.Comparing with the Uyghur automatic summarization based on semantic string technology,the experimental results demonstrate the superior performance of ontology technology on automatic summarization.In the ontology-based automatic summarization system,the semantic analysis of keywords is mapping vocabulary features into conceptual features,which improves the naturalness and coherence of abstracts.The extracted abstract sentence is more precise and more compact.