Combination of Big and Small Data: Empirical Study on the Distribution and Factors of Catering Space Popularity in Nanjing, China

Abstract

Using big and small data in urban studies has its advantages and limitations. Combining both data types instead of focusing only on big data is necessary. We took Nanjing as a case study to analyze the popularity of catering space using big and small data collected from Dianping.com and other sources. First, using online comments, we built a new index system to evaluate the popularity of catering providers and their spatial characteristics on the basis of a big-data approach. Second, we sampled a small group of catering providers registered on the website to discuss popularity factors through a regression model, whose variables were calculated from website data and small-data sources. Finally, we found a low level of catering provider popularity in Nanjing with a spatial distribution showing features of central flow theory (clear hierarchy, grading influenced by mobility and popularity, scattered and independent development). Some of the indicators having a significant effect on the popularity of these catering providers were type, size, per capita consumption, dish taste, atmosphere, service, basic price of commercial land, and number of similar providers within 300 m. Our study found a path for both combining big and small data and supporting catering space planning by urban governments, operation and site selection by catering providers, and dining and travel options chosen by consumers.