The Design Of The Model Of Natural Language Processing And Intelligent Search Engine

The methods for Natural Language Processing and current development are presented in the thesis. Auto-segmentation of Chinese vocabulary is researched. Several methods for segmentation of vocabulary are compared in the thesis. Through improving MM method, the efficiency of segmentation is enhanced. Table structure of segmentation and syntax theory are researched particularly. Through distinguishing all kinds of grammars between syntax theories, we find that type II grammar is most fit for natural language describing. And correlative arithmetic applied for analyzing syntax is given in the thesis, too.An Intelligent Search Engine Model, whose core technology is natural language processing, and its architecture are put forward later. The model is made up of Information Database, Natural Language Processor, Index Database and User Interface, etc.. Especially, function for all parts and arithmetic depicting are given. Its intelligence is embodied by maturity of knowledge and efficiency of arithmetic. Comparing to traditional engine technology, the model uses Notion Expanding to enhance recall, considers the importance of location of notion to improve precision, introduces dynamic building technology of searches vector to avoid data sparseness and to improve system efficiency. At the same time, in order to judge client's interest and answer automatically, the model applies for individual processing modules to sense client's request, and introduces modifying module for request to complete user-computer interaction.