A Study Of Chinese Preposition Disambiguation Combined Of Statistical And Semantic And Its Application In Mechanical Product Design

With the rapid development of computer hardware and software level as well as artificial intelligence, expert system are widely used in the production design, now almost all walks of life are the intelligent control of the computer integration into, so are the design of mechanical products. This paper attempts to understand the demand for automatic analysis of mechanical product design by using natural language understanding technology, through understanding, analysis and disambiguation the preposition structure expressed in the form of natural language in user’s needs, and the results of the analysis into the conceptual design requirements, providing support for the follow-up product design.The research status and characteristics of mechanical product design in the field of natural language understanding are briefly introduced in this paper, through analyzing the difficulties of the study of the semantic analysis and statistical analysis in natural language processing and comparison of several semantic analysis of the current methods and syntax analysis methods, the final determination of C hinese preposition disambiguation model of the combination of the statistical with semantic with the conceptual dependency theory as the theoretical basis.The difficulty of preposition structure analysis is how to classify prepositions and prepositional structure, mining the semantic information of preposition structure and effectively disambiguation processing, to solve this problem combined with the characteristics of natural language in the field of conceptual design of mechanical products, this paper firstly introduce the overview in studies the current situation of preposition structure and word sense disambiguation, and then respectively analyzes the semantic and statistical word sense disambiguation method, based on this, the author puts forward a kind of Chinese preposition disambiguation strategies combined with statistical and semantic. Semantic mainly in knowledge acquisition, knowledge representation method as the foundation, according to the classification Chinese prepositions, to construct a system of knowledge; At the same time, through the conceptual dependency tree model combined with the conceptual dependency theory,analyze the relation of the abstract level of C hinese preposition. Statistics is mainly in certain scale corpus, summarizes all kinds of C hinese prepositions collocation, by calculating the mutual information, analysis the probabilistic collocation between preposition and its collocation, and ultimately determine the results of the analysis.Finally, the Chinese preposition eliminating disambiguation analysis model combining semantic and statistical is applied to the natural language understanding system for conceptual design of mechanical products, by understanding and analysis of the user requirements, changing the results of the analysis into the conceptual design requirements, aiding the design personnel to conduct the follow-up design assistant. After debugging, semantic analysis programs are written achieved certain results.