Research On Social Sensing And Spatiotemporal Pattern Of Xiong’an New District Based On Weibo Data

In April 2017,China announced the establishment of the Xiong’an New District.The construction of Xiong’an New District has a great impact on China’s development.The public’s comments on the Xiong’an New District are important data to test the development effect of the New Area and find problems timely.The current research methods of the Xiong’an New District mainly focus on policy interpretation and field investigation.UGC(User Generated Content)data is an important source of public opinion monitoring.And Weibo,as one of the sources of UGC data,is an important platform for the public to release information.This article uses Sina Weibo as a data source,through multi-dimensional and comprehensive mining of the data,and find its spatial-temporal distribution,semantics and its spatial-temporal distribution,it discovers trends,summarizes laws,and raises questions to provide a reference for the construction of Xiong’an New District.This article uses Sina Weibo,which contains the keyword "Xiong’an New District" as the data source.From China,Beijing,Tianjin,Hebei,Baoding,and Xiong’an New District,from macro to micro scales,the data for two years in April 2019 was mining deeply: Firstly,the basic database was established,and the basic database included microblogging data and geospatial data after data cleaning;Secondly,the text mining library was establish.Text mining calculations includes TextRank keyword calculation,Bosonnlp sentiment calculation and LDA topic mining.Thirdly,the article visualized calculation and analysis based on the basic library and text mining library,Finally,through the interpretation of each category of the topic classification,count the topics’ popularity,and analyze it in time and space,the article excavate hot topics of public concern deeply,and finally built a semantic space-time visualization library.Research indicates:(1)Public attention is mainly driven by distance,economic development,population and information accessibility;Through hierarchical map visualization of the proportion of negative emotions,it can effectively monitor the anomaly space and mine negative events;The combined method of word cloud and chart visualization can dig out the hot content of public interest from a macro perspective effectively.(2)The time evolution of the heat in Xiong’an New District conforms to the distribution characteristic of major public opinion.In the beginning,the number of data broke out,and then the amount of data fell rapidly,showing a long-tailed distribution.The emotions are mainly positive,and the proportion of negative emotions in Baoding is relatively high.And word timing analysis can effectively mine social hot events.(3)A spatial pattern of high-value clusters in Beijing,Tianjin,and Hebei,low-value clusters in the west and northeast China,and central and southeast coasts in China has formed;The negative sentiment ratio has formed a spatial pattern of high-value clusters in the east China and low-value clusters in the west China.(4)Except for Baoding City,the changing rate of weibo number is in an “N” shape.The higher the rate of change,the greater the changing degree of attention this month.A high rate of change can reveal high-profile hotspot events;On a national scale,public attention is relatively stable with little change.However,in Beijing-Tianjin-Hebei scale,Hebei scale,and Baoding scale,the public attention content changes relatively frequently,and the word change line have upward trends,which indicates that the public attention is gradually dispersing.(5)The content of public attention can be divided into 25 second-level topics and 7 first-level topics.The first-level topics are external perspectives,infrastructure,functional positioning,real estate,government work,green livability and smart cities.Among the second-level topics,the topic of planning and construction is the most concerned.(6)By digging the trends of various topics,we can find the changing laws of the hot topics of public concern.For example,at Baoding,the attention of Baiyangdian Ecology and new model of innovation and development is increasing,which shows that Baoding attaches great importance to the innovation of Xiong’an New District and the protection of sustainable ecological development;By visualizing each topic’s heat in space,you can intuitively find the heat of each topic in each city.The cities were divided into three types: hotspot type,prominent type and zero type.By constructing multi-scale temporal and spatial heat and pattern calculation methods,using topic model and space-time analysis methods,this paper forms a complete,orderly and scalable social sensing model.Applying this model to the Xiong’an New District can effectively show the spatiotemporal law of Xiong’an New District’s attention,and excavate hot topics of public attention.It provides references for the construction of smart cities in Xiong’an New District,public opinion guidance and policy formulation in the New Area.It provides a reference model for the social sensing of other hot events.At the same time,it also provides new ideas for the development of media geography.