Research On Data Processing And Analysis System Based On Earthqueke Social Information Perception

In modern society,the research on natural science is gradually deepened,among which,the research on seismology has evolved from single observation to earthquake resistance,earthquake prevention,etc.,making full use of the social resource information of earthquake situation,reasonably adopting new methods to assist the government and earthquake departments to improve their quick response ability in the event of sudden earthquake,effectively avoiding the post earthquake risk,and providing effective services to the public The importance of relevant information and emergency rescue is particularly prominent.More and more people are publishing information in community forums,microblogs and other information websites,and a large number of social resources information of earthquake information are distributed in the above information websites and users' intelligent mobile terminals.In this context,it is of great auxiliary significance to study the sensing method of social resources information of earthquake situation and to process and analyze the data.In this paper,we design and develop an information awareness program,a set of information collection system,and data interaction between the mobile terminal and the server.According to the characteristics of different sources and different structures of earthquake information,this paper designs a set of coding and decoding rules to encode multi-source heterogeneous information.Gbdt and LR algorithm are combined to evaluate the seismic damage level.The evaluation effect of improving the original evaluation algorithm by using feature fusion technology and information coding technology is studied.In the last chapter,we test the function and performance of the whole system.The main research work of this paper is summarized as follows:(1)A network information perception program is developed,aiming at the community forum and the official information publishing website,the post earthquake disaster information perception is carried out.Based on the location and time of release,the information is classified and integrated.(2)The client and server of the system are designed.An android app is developed to obtain the information of the communication network related to the earthquake disaster of the intelligent mobile terminal.A server system is developed,which provides a platform for information reporting,realizes the function of information collection and persistence,and integrates and stores the information related to post earthquake disaster.(3)The data set is labeled,the feature is constructed,the evaluation model of earthquake damage degree is constructed,the combined feature vector is obtained by combining the extended feature of gbdt with the original feature vector,the data set is divided into test set and training set,the training set is used for LR model classification training,the prediction model is obtained,the test set is used for model prediction,and the performance of the model is compared according to the results Optimization.(4)A set of coding and decoding rules is designed for multi-source and heterogeneous post earthquake disaster data,which encodes the data with variable length and realizes the standard storage of heterogeneous data.Based on restful protocol,the interface of data acquisition and data collection is provided,and all interfaces are tested for function and performance.