The Study Of Intelligence Processing Method Based On Text Mining Technology

All along,the main function of intelligence agencies is to collect a large number of information materials collected to be analyzed and researched,and those information materials are converted into intelligence products and provided to the information users and support the user's decision-making behavior.In the various forms of information materials processed by the intelligence agencies,the text source of the open source is the main part of the information material being processed.With the rapid development of Internet technology in recent years,the amount of text information carried by online electronic documents,e-mail,database and so on has become the trend of explosive growth,and the source of information material is very rich.So,the traditional method of information processing is more and more difficult to meet the task needs.At the same time,intelligence users demand more and more information,and aging requirements are getting higher and higher.So,how to extract useful information from the vast Internet public information and improve the efficiency of processing and reporting has become an urgent need for the current intelligence agencies to solve the important issue.In view of the current intelligence agencies facing intelligence quality and intelligence efficiency of the two real problems,this paper attempts to use the technology of text mining to research preprocessing,classification and value mining of intelligence material,and to explore effective ways to improve intelligence processing efficiency and find out intelligence value from scattered intelligence material.The main work of this paper are:First,the construction of specialized intelligence field dictionary.Intelligence word segmentation is one of the important tasks of intelligence preprocessing.Based on the actual needs of intelligence processing,this paper puts forward the word segmentation strategy of intelligence dictionary based on specialized field,which is applied to the field of intelligence science,and constructs a simple dictionary of intelligence field to improve the accuracy of word segmentation.Second,improving the intelligence classification and designing system baced KNN algorithm.In this paper,KNN classification algorithm is used to classify the intelligence text.This paper analyzes the shortcomings of KNN classification method,and uses the improved KNN method to design the classification system.Through comparison experiments,it is found that the improved KNN classification method has improved accuracy and efficiency before improvement.Thirdly,the association rule algorithm and the intelligence analysis could be combined.We can use the association rule algorithm to derive the strong association rule between the elements of the military intelligence,and use the Apriori algorithm to excavate the depth of the information value.In view of the characteristics of military intelligence analysis,this paper adopts the optimized Apriori algorithm to focus on the importance of information elements,give greater importance to the important elements of information,give less important elements to less important elements.This paper also attempts to study the automatic generation of intelligence reports based on the location relevance of sea and air military targets.The research work and research results of this paper have some theoretical significance for the research of text mining technology in the field of information practice.In particular,combining the text mining technology with the actual information workflow can improve the efficiency of intelligence processing,and the effectiveness of the work of information processing has a reference value.