The Internet Public Opinion Models Under Uncertain Environment

With the rapid development of the internet,human society has marched into the network age.With the convenience bringing to people,the internet also brings a certain negative impact to human life,and the network public opinion has become a hot issue studied by many scholars in recent years.The characteristics of network public opinion is fast spread,short duration,and little historical data can be found.Therefore,the network public opinion only can be dealt with uncertainty theory at the present stage.This thesis proposes an internet public opinion model under uncertain programming first in the third chapter.Jumping out the imprisonment of differential equation,and trying to establish the network public opinion model from the optimization perspective.For solving the proposed model,this chapter designs a hybrid intelligent algorithm consisted by 99-method and an improved genetic algorithm.Finally,a numerical example is given based on the event of delisting 5A scenic spot of Shanhaiguan,Hebei,and the result shows good modeling and control performance.Moreover,since the uncertain factor influencing the internet public opinion is not single,three types of network public opinion under uncertain partial differential equation are established,that is soaring model,stationary model and recession model.Meanwhile,this thesis first proposes two methods of solving the first order linear uncertain partial differential equation and the inverse uncertainty distribution of solution is given too.The contributions of this thesis are:(1)Proposing the network public opinion model under uncertain programming for the first time.(2)Designing a hybrid intelligent algorithm to solve the proposed model.(3)Obtaining two methods of solving the first order linear uncertain partial differential equation,and giving the inverse uncertainty distribution of solution.(4)Proposing three types of network public opinion under uncertain partial differential equation.