Research On Modeling Method Of Evolutionary Trend Of Network Event Group Based On Difference Degree

Under the background of the information age,the way people obtain information is more and more diversified.With the continuous development of network technology,the Internet has become one of the important ways for people to obtain information.The dissemination of cyber incidents is fast,interactive and open.In the event development cycle,the public's high concern can lead to social panic or negative impact on social harmony.Therefore,in recent years,the party and the country's research and supervision on Internet public opinion has also begun.highly valued.The current research method of network event popularity mainly uses text processing methods to classify information on news websites or social forums and posts,thereby analyzing individual events.However,in today's information age,the correlation between events and events is ubiquitous.Events can be related to each other through common factors such as common subjects,topics,or emotions.It can be assumed that multiple events with the same or similar event prototype can constitute an event.Groups,they will be strengthened and highlighted by the occurrence of the same or similar events.Therefore,the goal of this paper is to study the heat evolution of network hotspot event groups,analyze the evolutionary trend relationships among related events,and find the evolution trend results of event groups.When similar events occur,we can judge the evolution trend of event groups.The evolutionary trend of events provides a theoretical and practical basis for the subsequent public opinion forecasting and warning.The analysis of the evolution of event groups can also provide some references for researchers of human behavioral dynamics.According to the existing research background and related technologies,this paper mainly designs and implements the following two parts:(1)In the first part of this paper,the hot news group news text is first crawled on the network.After the information is preprocessed,the single event is represented by the semantic fingerprint of the event,and the heat evolution of a single event over time is calculated.Then,the evolution diagrams of multiple correlated events are projected into the same timeline space and the proposed method is used to calculate the difference degree of the curves based on the event similarity.The matrix center is continuously updated,and the objective function is minimized to find the event group with the smallest difference degree of Evolution curve.Finally,the results obtained by this paper are compared with the results of the arithmetic mean and the similarity weighted avalue obtained in this paper is obviously better than the other two,which proves the verage of the events.It is found that the F calculation formula of the degree of difference based on the event similarity proposed in this paper.It is effective,so the evolution of the event group derived from the method presented in this paper can better reflect the uniform development characteristics of each event.(2)The second part of this paper combines the latest web development related technologies to implement a network event group evolution management system,and introduces the functions and implementation of the module such as information collection,information preprocessing,event group analysis,graph display,and system management.