| At present,our country is in a period of rapid economic development.Along with the advancement of urbanization process,which dues to various conflicts of interest,social problem increases prominently.Comprehensive Social Management(CSM)is put forward to maintaining the harmonious and stable society.With the help of information technology,CSM manage every field of social life into a unitive platform,such as the fields of political law cases,social order cases,petition investigation cases,livelihood services cases,urban management cases.The information management platform makes it easier and more efficient to manage the social life for the government.By using the data of CSM in Qingdao,this paper aims to study the method of data mining,which is used to mine the connotative information,such as the spatio-temporal information of livelihood services cases,carried by the dataset.So as to analyze the spatio-temporal correlation rule and spatio-temporal distribution characteristic of the CSM events.Finally,with the help of visualized technology,the information that discovered by data mining is displayed in a comprehensible way.The discovered information can be used in the field of decision-making in government's CSM work.And this paper is organized by the technical route in "Data analysis and preprocessing-information extraction from the data-mining the spatio-temporal information from the data".The main work of this paper is shown as follow.(1)The geographical features of CSM data and the analysis of the reliability of the data.In this paper,the geographical features of CSM data are analyzed elaborately.And the important attributes which can be used for data mining are summarized.The potential errors of the original CSM data are analyzed,and then the solutions of data error are given so as to make sure that the data is reliable.(2)Research on Information Extraction Method of CSM data Based on Ontology Semantic Reasoning.Considering that the information retrieval method can't understand the semantic information of the event text,this article uses ontology techniques to process and extract textual information.In this paper,by analyzing the concept system of CSM events,the ontology model of CSM events is constructed.Based on the ontology model,we designed the construction process of the social CSM event ontology.And then,an CSM event ontology is built.By using the Jena framework and the CSM event ontology file,we develop a system which supports ontology semantic reasoning.And with the help of the system,the accuracy of information extraction is improved.(3)Research on the spatio-temporal information mining method of social comprehensive management events.This paper introduces the spatio-temporal information mining for discrete CSM events from three perspectives:geographical statics,spatial aggregation and correlation relationship.Geographical statistics and spatial clustering exploration are based on the existing methods,which combined with the characteristics of social governance events carried out some improvements.Comprehensive relevance of the event to explore the method is the original method proposed in this article.(4)Research on spatial and temporal information mining and visualization method.Based on the ontology reasoning framework,the four types of events are extracted,such as heresy activities events,petition events,contradictory disputes events and brawl events.Digging their spatio-temporal distribution information,and show the result in a variety of visual means,such as heat map,scatter plot,line diagram.And then explain the result and make suggestions to the government's CSM works. |
PDF Full Text Request