Authors: Jimmy Feng*, University of Tennessee, Shih-Lung Shaw, University of Tennessee
Topics: Geographic Information Science and Systems, Food Systems, Spatial Analysis & Modeling
Keywords: food access, human perception, spatial accessibility
Session Type: Virtual Paper
Start / End Time: 8:00 AM / 9:15 AM
Room: Virtual 8
Presentation File: No File Uploaded
While there have been improvements to spatial accessibility measures to inform understanding of food access, many are still premised on the assumption that physical location is the most important determinant from both the supply and demand side. The fetishization of physical location may not accurately describe the situation of food accessibility and omits critical aspects related to how residents in an area obtain food and perceive their access to affordable and healthy options. Understanding food accessibility may be improved by not only looking at the quantity of grocery stores or the amount of time it takes to drive to a store but also by including human perceptions via online reviews. People may have different ratings (sentiments) and priorities (weights) for different aspects (semantics) of their food situation. With all this under consideration, this research explores the development of a mental-spatial accessibility measure to better understand accessibility to affordable and healthy/nutritious food locations. Reviews and ratings of grocery stores in Knox County, Tennessee from Google Maps are analyzed with Latent Aspect Rating Analysis (LARA). LARA uses a topic model to extract latent aspects and then decomposes overall ratings into individual aspect ratings. This can be used to understand people’s perceptions of the affordability and availability of healthy and nutritious food for different food retailers in the county, which complements the conventional measures of spatial accessibility.