Research on the impact of Dongguan manufacturing agglomeration on urban and rural residents' consumption upgrade based on deep learning

This study aims to use the convolutional neural network (CNN) model to conduct an in-depth analysis of the impact of manufacturing agglomeration on the consumption upgrade of urban and rural residents. Based on Dongguan's manufacturing enterprise distribution data, urban and rural residents' consumption behavior survey data, and agglomeration area economic data, this paper builds a deep learning model to capture the complex nonlinear relationship between agglomeration effects and residents' consumption behavior. The study found that manufacturing agglomeration plays a significant role in promoting residents' income growth and consumption upgrades. Especially in areas with high agglomeration, the probability of consumption upgrades increases significantly. Compared with traditional linear models, the CNN model automatically extracts the deep-level characteristics of regional manufacturing agglomeration on residents' consumption patterns through a multi-layer convolution structure, and demonstrates superior prediction accuracy. In addition, the research results provide data support for local governments to formulate differentiated policies to promote urban and rural consumption upgrades, especially in optimizing the industrial layout of agglomeration areas and narrowing the urban-rural consumption gap. However, this study also has certain limitations, including the breadth of data collection and model interpretability issues, which can be optimized and expanded in the future by combining more diverse data sources and more interpretable models.