Design And Implementation Of Hotspot Mining System Based On Internet News Flood Disaster

With global warming,the losses caused by meteorological disasters are on the rise.Meteorological disasters such as heavy rains,floods,high winds,thunder,hail and their secondary disasters show the characteristics of “various types of disasters,wide range of impacts,and severe impacts”.The occurrence of disasters has greatly affected people ’s lives.China has always been in a disaster-prone area,and once the disaster occurs,the losses caused are relatively serious.In view of the above,the thesis designs and implements a mining system for Internet news disaster hotspots.According to the theory and method of software engineering,first of all,the system needs analysis and overall design;then,the system is divided into five functional modules,namely the system management function module,the information extraction function module,the address management function module,and the noun database Management function module,hotspot display and data visualization function module.Users can retrieve webpage information related to disaster events through specific keyword combinations.The system also provides filtering and filtering functions.In addition,the system supports multiple ways of uploading geographical name databases,combined with Chinese Named Entity Recognition(NER)technology and optimization algorithms,which can manually identify and label geographic location information in web pages.For the management of address locations,users can view related information,including frequency,latitude and longitude,etc.,and perform information management operations.Finally,visualize all the acquired disaster-related information in various ways,such as map visualization and chart visualization.Then use Python language and Mongo DB database to design,develop and implement the system,and combine software testing methods to perform functional test and performance accuracy evaluation of the system.After system testing and analysis,it is shown that various types of information on the Internet,such as news media reports and user-generated data,can provide good support for search engines to obtain data.We hope this work will help further mining And explore Internet data resources,and develop an online,collaborative scientific analysis system for urban flood disaster management.In addition,the design and development of the system provides relatively complete ideas and processes for the acquisition and analysis of other types of public opinion events.The framework model of the system is also applicable to the analysis of other public opinion problems.