Design And Implementation Of Public Opinion Analysis System Based On Hadoop

With the rapid development of Internet technology,the information amount of network public opinion is increasing rapidly and the influence is increasing.It is very important to correctly grasp the trend of mass network public opinion to maintain the brand image of the enterprise.Collecting mass public opinion information from the network,excavating the positive and negative emotional tendencies of network public opinion,and providing public opinion statistics report can help enterprises to judge the reputation image of enterprises scientifically and maintain the brand image of enterprises in a timely manner.In view of the above situation,this paper puts forward the public opinion analysis system based on Hadoop,and uses the Hadoop platform to realize the storage,processing and emotional tendency statistics of mass network public opinion information,and provides an intelligent monitoring service for the image of the enterprise.Based on the research background and present situation of public opinion analysis system,this paper determines the introduction of Hadoop technology in the system,studies the network crawler technology,HTML DOM technology,emotion analysis technology and Hadoop technology,and lays a theoretical foundation for the realization of public opinion collection preprocessing analysis presentation.Demand analysis,determine the needs of the system,design the overall framework of the enterprise public opinion analysis system,and detailed description of the Hbase database table and field design,respectively,the system of public opinion collection module,web preprocessing module,public opinion statistics module and statistical fruit display module are designed in detail.It includes process and class design;the system has the ability to store and deal with large data.The system can be processed through MapReduce technology to realize the good visual interface of the system;the user starts from the requirement analysis of the system and tests the system comprehensively for the requirement analysis.By adopting a series of key technologies,this project has realized the public opinion analysis system based on Hadoop,designed the test case for the requirement of the system,tested the system comprehensively and tested the effectiveness and reliability of the system.The display interface of the system can be human-computer interaction and can show the emotional statistics of each enterprise.Top10 hot words,positive and negative emotional trends,attention trend,and so on.The pub lic opinion analysis system based on Hadoop supports the storage and processing of mass data,and can provide enterprise public opinion monitoring icon report for multiple enterprise monitoring.