Design And Implementation Of Public Safety Event Monitoring System In Microblog

With the rapid development of Internet technology and the continuous increase in the number of Chinese Internet users,social networks have changed people's traditional methods of information exchange,therefore more and more people have acquired and disseminated information through microblog.A large amount of event information is flooded in microblogs.Public safety event is what cause or have the potential to cause social harm.If we can monitor public safety event effectively,it can not only quickly alert public safety events,but also provide effective support for emergency management and decision-making.So it is very important to monitor public safety events from mass,complex and real-time microblog data on the Internet.Based on the characteristics of the propagation of public safety events in microblog,this paper studies a hot public safety event discovery method based on heat growth.First,the microblog data is divided according to the time period,and then the trend of microblog heat changes in the adjacent time period is calculated to determine the occurrence of public safety events.Further,after the public safety event occurs,the information related to the event is relatively complex and cannot provide the user with concise and effective event situation information.After monitoring public security incidents,this article first adopts the LDA-based public safety event description method to extract the main content of the event and provide a more intuitive description of the event.Then the analysis of public safety event based on Word2Vec's sentiment analysis method is used to provide the people's emotional tendency towards the event and to show the intuitive public opinion situation.Then we classify the people involved in the event.According to the emotional tendencies of different groups of people and the attributes of the people,identify the characteristics that can affect tendencies and formulate scientific public opinion guidance strategies.Finally,we use the microblog active user identification method to discover active users in public safety events.Through the active user's influence in events,we can better implement the public opinion guidance strategy.Based on the above technologies,this paper designed a public safety event monitoring and analysis system based on microblog.First,the web crawler obtains the microblog information,and then implements the system through data preprocessing,Chinese word segmentation technology,text representation technology,and text classification technology,combined with the public safety event discovery method based on the microblog data and the content-based public safety event analysis method.The system designed in this paper can effectively provide users with public safety event reminders,real-time status analysis of public safety events,public safety event details,public safety event descriptions,event-active user details,and event-related crowd details.