Fresh Retail Layout and Forecast Based on Multi-Source Big Data and Machine Learning: A Case Study of Guangzhou

Fresh retail is an important facility in the community life circle. Exploring its spatial distribution is essential for improving the allocation of life service resources. This paper introduces a method based on supply and demand matching and machine learning, designed to analyze and forecast the current state of supply and demand in offline fresh retail. Specifically, we take the four central districts of Guangzhou as the research object and apply the two-step mobile search method to filter the training set grid. Utilizing machine learning, we obtained the feature matrices of these grids and applied them to the supply prediction of the remaining grids. The results indicate that the Commercial Residential Density, Travel Convenience Index, and Food Service Density are the main factors affecting the location selection of offline fresh retail. The supply and demand in Yuexiu and Haizhu Districts show imbalance, and the construction of offline fresh retail facilities must be fortified.