Syntactic Analysis Of Kazakh Language Based On Chart Algorithm

Study on the levels of language can be divided into word segmentation, part of speech tagging, syntactic analysis, semantic analysis, textual analysis and information extraction. The syntactic analysis is the key position in the field of natural language research, and the accuracy of the results bring a serious impact on the further study. The purpose is to identify the parsing grammar relations including the word in a sentence and language between the blocks. The final results of the analysis show in the form of a data structure of the syntax tree.Firstly, the current research parsing algorithm for parsing algorithm has been controlled. Secondly, the two-pronged improved approach is introduced. First, the rule base for their activities proposed to eliminate the phenomenon of edge redundancy left recursive grammar improved algorithm; the second is the algorithm itself, which is the bottom-up method for adding activity chart analysis side optimization strategies. Examples of validation results show that:(1) eliminate the presence of left-recursive grammar rule left recursive algorithm can speed up the efficiency of the algorithm. When there is no left-recursive rule, this policy on the algorithm itself does not cause other effects;(2) the elimination of grammar left Recursive and activities edge optimization algorithm strategies simultaneously open at the time of both efficiency and accuracy than the traditional algorithm analysis Chart sentences were greatly improved; the superiority of performance,so selected it as the core of the syntax analysis prototype system.Again, standard selection of Kazakh syntactic tagset using Kazakh syntactic, and according to the usual sentence, designed a set of grammatical rules based on the context free grammar. Finally, based on the speech tagging module and Maximum Entropy Model Based on Improved Chart parsing module, designed to achieve an improvement in Kazakh chart analysis algorithm based on syntactic analysis system, the system is able to analyze a complete parse tree for a simple statement that the child Kazakhstan opened a new chapter, the next step will also improve the Kazakh language parsing system foundation, not only for Kazakhstan statement analysis method.Through the examples of testing, the improved algorithm play a better role, focus should be put back into the grammatical elements of grammar rules, grammar rules and continue to improve the size of the tree library, thereby improving the efficiency of Chart parsing system.