Understanding Of Key Technologies, Xml-based Operational Instruments

The operation document understanding was an important content in military automated command system. Key technologies of operation document understanding were researched in this paper. The main contents could be concluded as follows:(1) Because of the shortcomings of currently existing operation document, standardization for operation document with XML technology was realized in this paper: the Schema defined the logic frame of the operation document, the XML described the content of the operation document, and the XSL gave the format. Then according to the disposal flow of operation document, the architecture of the system was presented. The relation and conversion algorithm between the standardized operation document and the relation database was researched in the paper and the automatic management of the standardized operation was also implemented.(2) It analyzed the characteristics of Named Entities in the operation document. The method of Finite State Automaton (FSA) Matching was adopted to recognize several common types of Named Entities, such as time, location name, distance and army organization name.(3) A model of XML-standardized operation document understanding was designed, and the algorithms of the key modules in this model were researched in this paper. It studied the improved RMM Chinese word segmentation algorithm and part-of-speech tagging algorithm using Hidden Markov Model (HMM) and realized the segmentation and tagging of unknown words in the operation document. Facing the difficult problems of traditional Chart algorithm for syntactic analysis, such as ambiguation and efficiency, we proposed and realized an improved method for Chart algorithm which was named "score marks for rule".Experiments demonstrated that all the methods in the paper were effective.