Geographic Position Text Knowledge Mining Based On Syntactic Rules

People are faced with massive highly increased text data resources,along with the building of the internet technology,information technology and some other sorts of digital information.It has been a very important subject regarding how to make effective use of these massive text data and dig out all kinds of information and knowledge people needs.Geographic position text knowledge,a kind of declarative statement used to describe geographic position,is of great importance in advancing national information construction and defense construction.This paper takes multi-source geographic position text data as the research subject by using text mining technology and to get an intelligent mining on geographic position text knowledge based on the syntactic rules.Thus it is of research significance and application value.The following are the main research contents:1.According to the utilization of the geographic position text data,this paper states the research background and significance.Starting from the syntactic analysis of natural language,this paper introduces the research situation of Chinese syntactic rules.By the backtracking of the development of data mining technology,it analyses the current research situation in C hina as well as abroad.It also asserts the major research content and organizational framework.2.The paper analyses and summarizes the concepts,development and essential features.Combined with the concepts of geographic position,it outlined its text knowledge.The paper summarizes the characteristics of geographic position text knowledge.With the general procedure of text mining,it comes with the geographic position text knowledge mining model based on the syntactic rules,it builds the technological scheme.3.It describes the multi-source text data access to geographic position.It also analyses the text knowledge presentation method.On the basis of the relation between text knowledge and rule,as well as Chinese syntactic structure,it puts forward a new knowledge regularization construction method,and completes the corresponding construction of rule templates element,syntactic rules and rule tree.It states the text pretreatment concept and major function.It has studied a corresponding text pretreatment method dealing with geographic position text mining.A target text storage structure has been designed.4.It presents the pattern matching concept and utilization field and analyses the current improvement of pattern matching algorithm.It gives a detailed description of two classical single pattern matching algorithms KMP and BM functional idea and execution step.With the advantages of these algorithms,it put forward an improved single pattern matching algorithm.Its performance has been tested and analyzed.Using the improved method to complete the pattern matching on the basis of syntactic rules to get the knowledge mining,on which human-machine inter-matching is introduced.By referring blind comprehensive evaluating methods and hierarchy methods,it brings up an evaluation method.By calculating mining results comprehensive weight,it designs two different knowledge expression modes for the evaluating results.5.By applying these results to analyze the examples,it fully shows the technical process of data preparation,text pretreatment,text knowledge mining,mining results evaluating and knowledge expression.Thus fully verifies the precious theoretical research results.