Previous research on food geography mainly focused on analyzing how different types of food or ingredients are consumed in various places but often lacked explanations for the underlying reasons behind these patterns. In addition, some methods use statistical techniques to analyze regional dietary similarities. However, with the advent of the big data era, these traditional methods struggle to handle complex, high-dimensional data. To address these issues, we built a graph dataset, the Provincial Cuisine Preference Graph (PCPG), representing food preferences in each of the 31 provinces of mainland China from 2012 to 2022, using point of interest (POI) source data. Then, we quantitatively analyzed the distribution of various cuisines in different regions, followed by regression analysis to examine the impact of geography, economy, and population migration on regional food culture preferences. Furthermore, we proposed a framework, Graph-based Cuisine Clustering (GCC), for studying regional dietary similarities based on large-scale dynamic graph data. Our approach provides a detailed understanding of the factors influencing regional food preferences and offers a robust framework for future research in the field of food geography. Our code and dataset are available at https://github.com/trinitr0toluene/Graph-based-Cuisine-Clustering .

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Regional Food Culture Preference Mining Based on Restaurant POI

  • Ziyi Zhang,
  • Xu Chen,
  • Hao Huang,
  • Guojia Wan,
  • Xuan Zhou,
  • Shuang Li,
  • Yuhao Lin,
  • Jiawei Jiang

摘要

Previous research on food geography mainly focused on analyzing how different types of food or ingredients are consumed in various places but often lacked explanations for the underlying reasons behind these patterns. In addition, some methods use statistical techniques to analyze regional dietary similarities. However, with the advent of the big data era, these traditional methods struggle to handle complex, high-dimensional data. To address these issues, we built a graph dataset, the Provincial Cuisine Preference Graph (PCPG), representing food preferences in each of the 31 provinces of mainland China from 2012 to 2022, using point of interest (POI) source data. Then, we quantitatively analyzed the distribution of various cuisines in different regions, followed by regression analysis to examine the impact of geography, economy, and population migration on regional food culture preferences. Furthermore, we proposed a framework, Graph-based Cuisine Clustering (GCC), for studying regional dietary similarities based on large-scale dynamic graph data. Our approach provides a detailed understanding of the factors influencing regional food preferences and offers a robust framework for future research in the field of food geography. Our code and dataset are available at https://github.com/trinitr0toluene/Graph-based-Cuisine-Clustering .