Research On Public Opinion Monitoring System For Real-time Interactive Visualization

With the rapid development of Internet media,especially social media,every day there will be hundreds of millions of Internet users are active in social media.People can through the network media swift communication and sharing of information resources,so as to realize a wide range of communication.The audience at the same time receive information in the transmission of information,so the network media become the main way of people to express public opinion,communication feelings,is also the carrier of network public opinion formation and propagation.Network public opinion has become an important window for governments at all levels to understand public opinion,so the construction of network public opinion monitoring system is very important.In the network public opinion monitoring system,once the data delivery,must be presented to the user.In the face of a huge number of network information,how to efficient real-time data processing and visualization of the network public opinion has become a thorny problem.Static data visualization technology has been relatively mature,real-time streaming data visualization technology has been a research hot spot in the visual.This paper mainly aims at the real-time visualization of network public opinion system technology research,mainly includes two points:The firs hoe to real-time efficient processing network public opinion;the second how to network public opinion information real-time display to the user.A brief description of the background for the network public opinion monitoring system using Storm to network public opinion information real-time processing,the front-end of network public opinion through ECharts information real-time display.In this paper,the main work is as follows:(1)Improve Storm the default task scheduling mechanism.In view of the uncertainty stimulating factor alpha and beta,the stimulating factor alpha and beta in the Storm have been determined optimum value range of task scheduling,thereby increasing algorithm reasonable search directivity;For volatile factor of rho improvement,through introducing the Sigmoid function,make with the change of the number of iterations volatile factor rho adaptive change to speed up the convergence of the algorithm(2)Improve Storm the default task scheduling mechanism.By using the advantages of ant colony algorithm on NP-hard problem combined with the topology characteristics of Storm itself,an improved scheme of improved ant colony algorithm in Storm task scheduling is proposed.Reduces the load of each node in the CPU,and at the same time improve the load balance between each node,to speed up the efficiency of task scheduling(3)Design and implement public opinion monitoring and real-time visualization system Background from the demand analysis,general design,detailed design,system development of the network public opinion monitoring system in detail,fully in accordance with the development of the Internet project steps to carry on the design in this paperExperimental verification,the Storm,the improved ant colony algorithm is the default round-robin scheduling algorithm in CPU load was reduced by 26%on average,CPU use standard deviation was reduced by 3.5%at the same time,the algorithm efficiency than the Storm the default round-robin scheduling algorithm was improved by 22.6%.Improved ant colony algorithm,the Storm of task scheduling is superior to Storm the default task scheduling Design a kind of network public opinion real-time streaming data visualization scheme,implements the network real-time computation,storage and display of public opinion.This system combine real-time processing and visualization,enables decision makers to grasp the development trend of network public opinion in the first time and understand the attitude of the audience.The experiment confirmed the practical feasibility of the system to achieve the project expected design requirements.