Font Size: a A A

Research On Spatial-temporal Characteristics For Hot Spot Of City Based On Location Check-in And POI

Posted on:2019-02-26Degree:MasterType:Thesis
Country:ChinaCandidate:X Y XuFull Text:PDF
GTID:2370330545971195Subject:Surveying the science and technology
Abstract/Summary:PDF Full Text Request
With the rapid development of the Internet,the popularity of smart mobile devices,the widespread use of positioning systems,there are more and more applications based on location services,the number of users has developed rapidly.The number of users of smart mobile devices is increasing.They use location services to store their own location information or information corresponding to location,and eventually gather a large amount of user location data.Using data mining methods to explore vast amounts of check-in data,we can obtain basic information such as gathering points,object distributions,and urban hotspot.Then,using spatial analysis to explore the relationship between sign-in data,roads and POIs.We can provide a theoretical basis for the smart design of cities and the rational allocation of urban resources by exploring the status quo of urban development and the characteristics of spatial and temporal distribution.This paper proposes the DBSCAN clustering algorithm based on the shortest path and studies the temporal and spatial characteristics of hot spots in Jinan based on location check-in and POI by using Python's Scrapy framework to crawl Sina Weibo's check-in data.The main research contents and achievements are as follows:(1)Acquisition of Sina Weibo user data.Nowadays,due to Python's Scrapy framework is open source and free use,it is widely used for network data acquisition.We can analyze the meaning of each field of the acquired data and use the Python to write program to do appropriate processing.Then we can obtain the POI of the microblog by using the crawler and write it into Oracle,and connect the microblog posted by the user with the POI to form the POI check-in data.(2)This paper proposes a DBSCAN clustering algorithm based on the shortest path,and uses this algorithm to analyze the POI registration data of sina micro-blog users in Ji'nan.The study found that the greater the intensity of road and POI,the higher the number of users sign in,the sign of the sign in a lot on both sides of the road or road,the population is often concentrated in these areas.Then through the frequency of each POI sign to do the buffer analysis,determine the main activity space in each area,analyze the living space around education space,commercial space,and show the spatial distribution characteristics of Ji'nan.(3)We found that the flattening rate of the check-in ellipse in October 2016 was significantly larger than that in November 2014,which that compared to Jinan in 2014 in 2016 Urban are continuously expanding in recent years,and the trend toward eastward development is particularly evident by using the standard deviation ellipse to analyze POI check-in data for 2014-11 and 2016-10.From the perspective of the standard deviation ellipse,the 16-year area was nearly four times that of 14 years,and the extra area occupied nearly one-half of the eastern area.It indicates that the eastern part of Jinan Cith like Gaoxin District,Licheng District,etc.These regions has attracted a large number of talents,and they use location check-in services to leave footprints on the map,and the direction of talent flow further explains the trend of Jinan's eastward development.
Keywords/Search Tags:Sign in Data, Cluster Analysis, Shortest Path, Data Mining
PDF Full Text Request
Related items