Extracting urban functional regions from points of interest and human activities on location‐based social networks

Abstract

Data about points of interest (POI) have been widely used in studying urban land use types and for sensing human behavior. However, it is difficult to quantify the correct mix or the spatial relations among different POI types indicative of specific urban functions. In this research, we develop a statistical framework to help discover semantically meaningful topics and functional regions based on the co‐occurrence patterns of POI types. The framework applies the latent Dirichlet allocation (LDA) topic modeling technique and incorporates user check‐in activities on location‐based social networks. Using a large corpus of about 100,000 Foursquare venues and user check‐in behavior in the 10 most populated urban areas of the US, we demonstrate the effectiveness of our proposed methodology by identifying distinctive types of latent topics and, further, by extracting urban functional regions using K‐means clustering and Delaunay triangulation spatial constraints clustering. We show that a region can support multiple functions but with different probabilities, while the same type of functional region can span multiple geographically non‐adjacent locations. Since each region can be modeled as a vector consisting of multinomial topic distributions, similar regions with regard to their thematic topic signatures can be identified. Compared with remote sensing images which mainly uncover the physical landscape of urban environments, our popularity‐based POI topic modeling approach can be seen as a complementary social sensing view on urban space based on human activities.

Using Spatial Semantics and Interactions to Identify Urban Functional Regions

“…The different types of POI can be located in urban functional regions to support human activities. POI data provide information about urban functional zones, which is based on human cognition with spatial, temporal and semantic granularity [22][23][24][25]. POI data often describe the physical location and attribute information of places, such as residential buildings, leisure parks, commercial points and public services.…”

“…The spatial distribution and geographic knowledge of POI categories can be employed as spatial semantics to annotate functional regions [24,29]. However, few studies have simultaneously considered both spatial semantics and spatial interactions to identify urban functional regions.…”

The Combined Use of Remote Sensing and Social Sensing Data in Fine-Grained Urban Land Use Mapping: A Case Study in Beijing, China

“…The combination of the two data sources takes into consideration both the physical and socioeconomic attributes of a field, and thus is expected to hold great potential for providing better insights into urban landscape patterns and for mapping urban land use more accurately [20]. However, most urban land use studies relies solely on either remote sensing data [2,21] or social sensing data [22][23][24][25], but seldom on both [20]. Consequently, the importance of attributes derived from both data sources for use in classifying different urban land use types is rarely addressed in the literature.…”

A dynamic human activity‐driven model for mixed land use evaluation using social media data

“…Most of the current research directly adopted static land use classes extracted from existing nomenclatures (Gehrke & Clifton, ), which were laboriously obtained through either remote sensing images or direct observations (Heiden et al, ), to unveil physical urban structures. However, the dynamic human activity information will be lost when extracting mixed land use indicators, which are highly relevant to the functions of places (Gao, Janowicz, & Couclelis, ) and can have great influence on land use diversity.…”