Analysis Of Urban Hotspots And Commercial Area Exploration Based On User Access Logs From Public Map

With the rapid economic growth,the phenomenon that many domestic urban commercial areas sprawl is getting worse after the reform and opening up.However,However,our current research and understanding on urban commercial area is less,which is not conducive to the planning department to make reasonable guidance.The progress of Internet technology makes the network public map service has been rapidly popular.The log data generated by the user access to the public map website is a kind of geography data which has rich geographical location information,and it is of great value to study the Web log data.In this paper,a method of urban commercial area measurement based on public map access data is proposed to solve the problem of poor timeliness,high cost and low accuracy.In order to obtain high quality data,this paper chooses Hadoop and related components as the basic framework of data preprocessing,stores the original data in the distributed file system(HDFS),extracts,clean,convert the data with ETL(Extract Transform Load)data processing method.The ETL data is stored in the HBase database for subsequent analysis.In this paper,we use grid method to transform the data into grid data,which has spatial continuity and can better reflect the density of access events.In order to excavate the hotspot region and the spatial distribution pattern of the commercial area,the global spatial autocorrelation test of the data is explored by exploratory spatial data analysis method to explore the spatial distribution pattern of public map access data.Then the optimal clustering distance of the research area is explored and the spatial relation matrix is constructed.The local spatial autocorrelation method is used to detect the hot spot region.Finally,the commercial high value hotspots are extracted and the standard deviation ellipses are used to measure the urban commercial area.In order to select the appropriate research area,the number of public map visitors in each city is counted,and the number of users in each city is found to be in accordance with the Zipf distribution.The results show that the distribution of Shenzhen commercial area based on public map access data is highly correlated with the commercial area planned in Shenzhen,and can be used to provide services for government planning for regional pianning and enterprise operation and management.