Empowering health geography research with location-based social media data: innovative food word expansion and energy density prediction via word embedding and machine learning.

BACKGROUND: The exponential growth of location-based social media (LBSM) data has ushered in novel prospects for investigating the urban food environment in health geography research. However, previous studies have primarily relied on word dictionaries with a limited number of food words and employed common-sense categorizations to determine the healthiness of those words. To enhance the analysis of the urban food environment using LBSM data, it is crucial to develop a more comprehensive list of food-related words. Within the context, this study delves into the exploration of expanding food-related words along with their associated energy densities. METHODS: This study addresses the aforementioned research gap by introducing a novel methodology for expanding the food-related word dictionary and predicting energy densities. Seed words are generated from official and crowdsourced food composition databases, and new food words are discovered by clustering food words within the word embedding space using the Gaussian mixture model. Machine learning models are employed to predict the energy density classifications of these food words based on their feature vectors. To ensure a thorough exploration of the prediction problem, ten widely used machine learning models are evaluated. RESULTS: The approach successfully expands the food-related word dictionary and accurately predicts food energy density (reaching 91.62%.). Through a comparison of the newly expanded dictionary with the initial seed words and an analysis of Yelp reviews in the city of Toronto, we observe significant improvements in identifying food words and gaining a deeper understanding of the food environment. CONCLUSIONS: This study proposes a novel method to expand food-related vocabulary and predict the food energy density based on machine learning and word embedding. This method makes a valuable contribution to building a more comprehensive list of food words that can be used in geography and public health studies by mining geotagged social media data.

Investigating the association between the socioeconomic environment of the service area and fast food visitation: A context-based crystal growth approach.

Fast food consumption is one of the major causes of rising obesity rates. Fast food consumers are mostly residents located in the service area-the fast food outlet's surrounding area. Conventional buffer approaches may exhibit bias in measuring service areas by ignoring the local community's detailed spatial configuration and transportation preferences. This study uses fast food outlets and their visits provided by a mobile phone-based dataset named SafeGraph and applies a novel context-based crystal growth algorithm (CG) to delineate improved service areas of fast food outlets in Chicago. We also explore how socioeconomic variables in service areas by CG and buffer-based approaches are related to visits to fast food outlets. Results show that compared to conventional buffers, CG produces improved service areas as it excludes inaccessible barriers and adjusts the accessible areas by transportation preferences. Further, the model using service areas of public transport users by CG yields the best performance. Additionally, the rate of single-mother households and the number of other fast food outlets nearby are positively related to fast food visits in all models. Findings acknowledge the advantages of CG and help make policy interventions to reduce fast food consumption.

Changes in accessibility to emergency and community food services during COVID-19 and implications for low income populations in Hamilton, Ontario.

In this paper we analyze the changes in accessibility to emergency and community food services before and during the COVID-19 pandemic in the City of Hamilton, Ontario. Many of these food services are the last line of support for households facing food insecurity; as such, their relevance cannot be ignored in the midst of the economic upheaval caused by the pandemic. Our analysis is based on the application of balanced floating catchment areas and concentrates on households with lower incomes (