Research On The Multimedia Network Public Opinion Critical Monitoring Index System Of Building

With the development of 4G network and the popularity of mobile terminals,public way of expressing emotion,published by the traditional medium of information transfer to multimedia Internet platform to promote multimedia network public opinion information into the public eye.The public through the APP software,smart phones,all kinds of mobile terminal portable,fast,to participate in the microblog,We Chat,BBS and other media of network public opinion information.Compared with the traditional media such as newspaper,TV,multimedia network public opinion has spread fast,huge information quantity,storage,management,analysis,difficulties,crisis of public opinion formation probability is high,the impact range,strong influence,etc.Once the formation of public opinion crisis,loss is immeasurable.Relevant departments timely monitoring and analysis of multimedia network public opinion crisis,as a precondition for guiding,directing public opinion crisis.The index system of building is the base of the crisis monitoring and the basis.The purpose of this paper is to,under the construction of multimedia network public opinion crisis monitoring index system.First of all,in the research of multimedia network public opinion propagation process and evolution analysis of the influence factors on the basis of this article refine covers five elements,the development of the public opinion,namely the object subject of public opinion,public opinion,public opinion ontology,medium of public opinion,public opinion environment,the index system of them as primary index,at the same time in the research literature at home and abroad and the index system of authority,set up the secondary indicators,and quantifiable level at the end of the index is given.The index system objectivity is stronger,has extensive representative;Secondly,in view of the existing multimedia network public opinion characteristics of large amount of information,this paper,by using big data processing platform,the use of clustering,classification,association rule data mining techniques such as dynamic data collection and rapid processing images,text,video and other semi-structured,structured and unstructured data;Again,the indexes selection,using correlation analysis,principal component analysis method.Using correlation analysis method,first remove the correlation coefficient is largerwithin the same layer index set of indicators,to ensure that the rest of the index system can represent the initial indicator set information,reduce the initial indicator set the degree of redundancy.Then using principal component analysis(pca),through linear transformation,the remaining indicators set dimension,and with the aid of the cumulative variance contribution thought to verify the final index system redundancy is low;Finally,BP neural network based on genetic algorithm is adopted to define the filter after the weight of the indexes at all levels,by using Delphi method and SPSS analysis method to set the initial control parameters and the combination of public opinion crisis level,both in the macro injection importance index system,and to determine the index weights on the micro objectivity.The issue of "Korean" nikzad.For instance model,validate this article build multimedia network public opinion crisis monitoring index system of the rationality and necessity.