A Visual Analysis System Of Human Spatial-temporal Semantic Mobility Pattern Based On Data Confusion

With the development of the city,the population in cities is becoming more and more intensive,which result in terrible traffic congestion,bad air pollution and other urban problems.Therefore,the study of human mobility pattern has been attached great importance to government workers,urban planners and daily travel citizen.The main challenge is that it is difficult to mine the temporal and spatial mobility pattern,and so is the intent and the semantic information of human mobility.Based on the similarity of multi-source data,this paper integrates vehicle bayonet data,Weibo data,and POI data of city,so that road vehicle bayonet data with abundant information and urban POI data can be supplemented into Weibo data.Finally,we can extract semantic information from enhanced semantic Weibo data by this way.This paper is aimed to study the temporal and spatial human semantic mobility pattern.Combined the time view that shows urban traffic hourly flow,with the word cloud view showing key words from enhanced Weibo data,the users could find the time period that they are interested in.After the user determines the time period to be analyzed,then analyses a radar view that shows the regions' attribute,a word cloud view that shows the key words in Sina Weibo,and a geographic view that shows human mobility from origin to destination.,and finally combine the Weibo list that shows the Weibo content while some key word in the key word cloud view is clicked.After that user can mine out not only the numerical but also semantic human mobility patternWe implement this visual analysis system of this paper based on B/S framework,the front-end is used for the display of various visual components,and the back-end processes data according to the parameters passed by the user.In order to speed up the processing of data in the back-end,we create Lucene index of vehicle bayonet data and Weibo data.Users can use this visual analysis system to mine crowd movement numerical and semantic patterns from global visualization to local visualization.