Environmental Protection Network Public Opinion Analysis And Management Strategy Research Based On Information Ecological Model

In recent years,China has proposed to build a nationwide action system for ecological environmental governance,encouraging the public to participate in environmental supervision by public opinion.In the current participation path,the most common form of public participation is to express opinions and suggestions through online platforms and carry out supervision by public opinion.Internet platforms have become an important source of public opinion on environmental protection in China.However,compared with traditional environmental public opinion,the openness and concealment of the Internet and its unique fission mode of information propagation enable environmental network public opinion to form a huge field of social public opinion in a short time,change the development trend of events to a certain extent,destroy the balance of information ecosystem,and even cause serious social harm.The theory of information ecology emphasizes the harmonious development among the elements of information ecosystem from the perspective of system.Therefore,how to effectively use the theory of information ecology to build an information ecological analysis model of environmental protection network public opinion and make a systematic analysis of environmental protection network public opinion is the effective entry point for the government to effectively control public opinion,maintain the stability of information ecosystem and smoothly promote the construction of ecological civilization.On the basis of reviewing and combing the literature at home and abroad,this paper comprehensively uses the research methods of literature analysis,simulation analysis,social network analysis and emotional analysis to construct the ecological analysis model of environmental protection network public opinion information from the three aspects of overall propagation law,social network characteristics and public emotional tendency,systematically analyzes the environmental protection network public opinion and puts forward management strategies.Specifically,this paper takes the "waste classification" network public opinion as a typical case to study the environmental protection network public opinion.Firstly,combine the information ecology theory with environmental protection network public opinion,and construct the environmental protection network public opinion information ecology analysis model;Secondly,based on the information environment elements,a logistic growth model is constructed to simulate and analyze the law of public opinion information propagation on the "waste classification" environmental protection network,and quantitatively divide different propagation stages;Thirdly,based on the information person element,the network structure and node characteristics of "waste classification" environmental protection network public opinion in different communication stages are compared and analyzed through social network analysis;Then,based on the information elements,the Bidirectional Encoder Representations from Transformers(BERT)-Convolutional Neural Networks(CNN)emotion analysis model is constructed to classify the public emotion of the key nodes in the "waste classification" environmental protection network public opinion social network,and the word frequency feature analysis and semantic network analysis are carried out on the basis of emotion classification;Finally,it returns to the information ecology theory and puts forward the management strategy of environmental protection network public opinion,which is the foothold of full-text management practice.The main conclusions of this paper are as follows:(1)For the analysis of the propagation law of environmental protection network public opinion,the quantitative stage division of environmental protection network public opinion can be carried out according to the logistic model.The growth rate of public opinion information on environmental protection network shows a trend of increasing first and then decreasing.The propagation stage can be divided into incubation period,outbreak period,repetition period and long tail period.And only the change of the growth rate will cause the change of the stage length of the outbreak and repetition period,so the early warning level of public opinion can be judged by the growth rate.The phenomenon of "multiple peaks" exists in online public opinion on environmental protection,which can cause the growth rate to change with the change of different peaks,and the growth rate of different stages is quite different from the overall growth rate.Therefore,it is necessary to conduct a fine-grained analysis for different life cycles.(2)For the analysis of the social network of environmental protection network public opinion,in the incubation period and long tail period,the popularity of public opinion propagation is low,the centrality of feature vector is low,but the clustering coefficient is high,and most of them are information propagation by the government and mainstream media;In the outbreak period,public opinion spread widely,the propagation link is long,there are more mainstream media,and the centrality of feature vector reaches the highest,but the clustering coefficient is the lowest;However,in the repetition period,a large number of "we media" participate in the discussion of public opinion topics,the government intervenes less and less,and the media’s position as opinion leader becomes increasingly prominent.Compared with the outbreak period,the feature vector centrality is lower but the clustering coefficient is higher.Therefore,guiding the mainstream media and we media to forward and discuss public opinion is of great significance to correctly control the guidance of public opinion and promote the development of China’s environmental protection work.(3)For the analysis of the text emotion of environmental protection network public opinion,this paper constructs an emotion analysis model based on the combination of BERT and CNN.After training,the accuracy rate is 88.5%.The model can accurately complete the task of text classification.Using this model to classify the "waste classification" network public opinion,the polarization of public emotion is obvious.43% of the public show negative emotion towards garbage classification and 46% show positive emotion towards garbage classification.From the positive emotion semantic network analysis,the publicity of environmental protection concept and waste classification policy can effectively promote the implementation and promotion of waste classification.From the negative emotion semantic network,we can see that the focus of public attention mainly focuses on the problems of limited delivery,non classified transportation of garbage trucks,non-uniform classification standards and so on.The main reasons for public dissatisfaction are insufficient publicity and education and public awareness of environmental protection,imperfect waste classification policies,laws and regulations,and backward infrastructure and collection,transportation and treatment methods.(4)The management strategy and mechanism of environmental protection network public opinion are put forward.This paper divides the management strategy into three aspects.In terms of information environment management strategy,this paper puts forward internal environment management strategy and external environment management strategy.In terms of information person management strategy,this paper puts forward opinion leader management strategy and ordinary netizen management strategy;In terms of information management strategy,this paper puts forward the emotional polarity management strategy.The management mechanism is also divided into three aspects: environmental protection network public opinion early warning mechanism,guidance mechanism and control mechanism.It is pointed out that we should start early warning in the latent stage of public opinion,guide in the outbreak and repetition stage,control in the process of public opinion development and deal with the aftermath.