A Data Mining Technology Research Based On Internet Public Opinion Of International Military Relationship

As an important part of international relations,military relations are an important proposition for research under the relevant disciplines of national strategy and national defense strategy.Due to the complexity of traditional inter-state relations and the limitations of processing techniques,the current research on bilateral relations mainly relies on semi-structured or structured data,which makes the data sources have certain limitations.At the same time,in the analysis process,semimanual or manual methods are used to investigate and analyze data through human expert experience.This manual approach consumes lots of human resources and often has timeliness problems.To date,motivated by computer science and automation technology,the universal application of the Internet has made a qualitative leap in the speed and volume of message and information transmission.In addition,the intelligent analysis technology for text and multimedia information has made the technical idea of data-driven decision making possible.Therefore,how to use the relevant intelligent technology to analyze the massive data of the Internet to complete the study and investigate of international military relations,so as to better serve national security and benefit the society is a very important research issue.Based on this issue,this paper designs and implements an international military relationship network platform based on Internet public opinion.The platform first uses the crawler to collect relevant text information in the Internet,and then through emotion analysis,media influence calculation,information credibility technology,entity extraction and integration research to mine the value of the collected information and transform it into structured knowledge.This paper builds an international military relationship network based on the relevant technology of graph theory,and conducts quantitative analysis and prediction of military relations between countries.Finally,based on the above research,this paper designs and implements a visualization prototype system of international military relationship network based on Internet public opinion.In summary,the main contributions of this article are as follows:1.Aiming at the multi-source media information of the Internet,this paper proposes a dynamic multi-information collection and processing framework.In detail,this paper firstly designs and implements structured information as well as deep network information acquisition technology collection for news media;Second,this paper calculates the sentiment tendencies on the above text information;Third,this paper conducts an impact analysis through the propagation and attention of text on the Internet,so as to make a weight estimation based on the impact analysis results;Fourth this paper builds a credibility recognition scheme for news and social media text information;finally,this paper proposes a high-performance solution based on recurrent neural networks for the particularity of military entity extraction and integration.2.For the multi-source information collected and processed,this paper proposes the construction and application technology of the international military relationship network.in particular,First,based on the collected and processed data,this paper constructs an entityrelationship network based on international military relations through graph theory related technology;Second,based on the constructed network,this paper conducts quantitative measurement and prediction of military relations between the two countries,and verifies it through actual historical data.3.The paper completes the technical design and implementation of international military relationship mining and analysis technology based on Internet public opinion.