Research Of Knowledge Query Algorithm Based On Natural Language

With the deepening of social development, human has a higher demand for the access to information and knowledge query. Under the demand a strong driven-force has been generated and there is a powerful incentive to depth theoretical research and technological innovation about the answering system, natural language query system as well as search engine by the majority of scientists. Among of them, the knowledge engineering and natural language processing and other fields play a vital role. The natural language query based on knowledge base system is a valuable area of research. The core of knowledge base system is the knowledge base as well as the core of knowledge base are the knowledge representation and knowledge acquisition. However, knowledge query algorithm is the core field in the natural language processing and the key of knowledge query algorithm are the segmentation algorithm and matching algorithm. Under the background, the author studies the paper about the research of knowledge query algorithm based on the natural language.Amounts of theory base mainly contain knowledge engineering, natural language processing, relation model and parallel computing and so on. In the context of this theory, the main innovations of the paper are the following:Firstly:Analyzing the method of semantic network knowledge representation and the strengths and weaknesses of the relational mode and putting forward a method of semantic network knowledge representation based on relation model.Secondly:Improving the forward maximum matching algorithm as well as proposing an improved forward maximum matching algorithm in the knowledge query algorithm.Thirdly:Proposing a sentence template storage structure based on the sentence parse tree in the knowledge query algorithm.Fourthly:Proposing a sentence template matching algorithm based on the sentence template storage structure, including filtering algorithms and tree matching algorithm.At last, put forward the optimizing solution for the existed problem on the knowledge query algorithm.Author study the paper about pure theory research based on natural language (Chinese text),but the thesis has also exposed some problems:First, the paper is pure Chinese character text-based natural language queries, and social needs of today tend to be based on Chinese characters, numbers, and other English text characters mixed query:Second, the paper is based on pure research, the proposed algorithm written only by the class in the form of pseudo-code, and no program designed to achieve the algorithm and its validation and testing.