Evaluating trade areas using social media data with a calibrated huff model

Yandong Wang, Wei Jiang, Senbao Liu, Xinyue Ye, Teng Wang

Research output: Contribution to journalArticlepeer-review

28 Scopus citations

Abstract

Delimitating trade areas is a major business concern. Today, mobile communication technologies make it possible to use social media data for this purpose. Few studies however, have focused on methods to extract suitable samples from social media data for trade area delimitation. In our case study, we divided Beijing into regular grid cells and extracted activity centers for each social media user. Ten sample sets were obtained by selecting users based on the retail agglomerations they visited and aggregating user activity centers to each grid cell. We calculated distance and visitation frequency attributes for each user and each grid cell. The distance value of a grid cell is the average distance of user activity centers in this grid cell to a retail agglomeration. The visitation frequency of a grid cell refers to the average count of visits to retail agglomerations by user activity centers for a cell. The calculated attribute values of 10 sets were input into a Huff model and the delimitated trade areas were evaluated. Results show that sets obtained by aggregating user activity centers have a better delimitating effect than sets obtained without aggregation. Differences in the distribution and intensity of trade areas also became apparent.

Original languageEnglish (US)
Article number638868205
JournalISPRS International Journal of Geo-Information
Volume5
Issue number7
DOIs
StatePublished - Jul 2016
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Geography, Planning and Development
  • Computers in Earth Sciences
  • Earth and Planetary Sciences (miscellaneous)

Keywords

  • Huff model
  • Social media
  • Spatial aggregation
  • Trade area
  • User selection

Fingerprint

Dive into the research topics of 'Evaluating trade areas using social media data with a calibrated huff model'. Together they form a unique fingerprint.

Cite this