Mapping Urban Spatial Structure Based on POI (Point of Interest) Data: A Case Study of the Central City of Lanzhou, China

Abstract

The study of urban spatial structure is currently one of the most popular research fields in urban geography. This study uses Lanzhou, one of the major cities in Northwest China, as a case area. Using the industry classification of POI data, the nearest-neighbor index, kernel density estimation, and location entropy are adopted to analyze the spatial clustering-discrete distribution characteristics of the overall economic geographical elements of the city center, the spatial distribution characteristics of the various industry elements, and the overall spatial structure characteristics of the city. All of these can provide a scientific reference for the sustainable optimization of urban space. The urban economic geographical elements generally present the distribution trend of center agglomeration. In respect of spatial distribution, the economic geographical elements in the central urban area of Lanzhou have obvious characteristics of central agglomeration. Many industrial elements have large-scale agglomeration centers, which have formed specialized functional areas. There is a clear "central-peripheral" difference distribution in space, with an obvious circular structure. Generally, tertiary industry is distributed in the central area, and secondary industry is distributed in the peripheral areas. In general, a strip-shaped urban spatial structure with a strong main center, weak subcenter and multiple groups is present. Improving the complexity of urban functional space is an important goal of spatial structure optimization. ISPRS Int. J. Geo-Inf. 2020, 9, 92 2 of 26 that influence the urban spatial structure, such as knowledge and information, which enhance the traditional physical production factors such as land and labor [4]. Due to the arrival of the era of new data, an increasing amount of urban research is carried out by using a wide variety of geoscience new data. The applied data mainly focus on items such as the map POI (point of interest) data that describe the geospatial entities, GPS track data, and mobile phone signaling data related to crowd activities [5][6][7][8][9][10][11][12]. POI data, as a new spatial data source, have an important application potential. In this paper, POI refers to some geographical entity, which is closely related to human life. And in a geographic information system, POI can be a commercial outlet, a bus stop, a high-rise building, etc., which are more authentic. Therefore, POI data can largely enhance the capacity of the description of the physical location, reflect city activities, and can be effectively applied to the urban spatial structure research [13]. Compared with traditional data, POI data not only have the advantages of a large sample and easy access, but can also represent the spatial distribution of social and economic activity intensity and functional composite utilization, and reflect more accurately urban spatial structure [14,15]. In this context, research on the combination of traditional urban space and geographic new data is in urgent need of empir...

Who Will Survive and Revive Undergoing the Epidemic

“…Existing works have provided valuable knowledge on POI visit modeling for epidemic prediction [7], urban planning [16,25], recommendation [33,36] and other important applications. Chang et al [7] leverage the POI visit records to facilitate epidemic modeling.…”

Identifying Urban Building Function by Integrating Remote Sensing Imagery and POI Data

“…KDE is a useful method to estimate unknown density function in probability theory. It computes the density contribution of each sample point in the specified range to the centre point of each grid through the kernel function, and finally generate a smooth surface indicating the density map [52]. The amount of POI data in different categories varies greatly, for example, the number of POI data in commercial and business category are greater than the number in other categories.…”

Comparison of Random Forest Model and Frequency Ratio Model for Landslide Susceptibility Mapping (LSM) in Yunyang County (Chongqing, China)

“…If each POI site is regarded as a functional unit, then the higher the POI density, the more concentrated the urban functions in an area. POI kernel density analysis was made with ArcGIS software, which is often used to identify urban centers, economic vitalities, and intensities of human activities [55][56][57]. However, in the field of LSM, no research has been found when using POI as an influencing factor in human engineering activities thus far.…”