Consumer behavior analysis and consumption upgrade demand mining based on DBSCAN clustering algorithm

As e-commerce platforms become more and more developed, consumer behavior analysis has gradually become a particularly important research direction, mainly to better optimize e-commerce marketing strategies. Because today's consumer behavior is too complex and too diverse, it actually puts higher requirements on traditional data analysis methods. How to accurately identify different consumer groups and how to tap into users' potential needs for “consumption upgrades” are actually very urgent problems to be solved. This paper introduces the DBSCAN clustering algorithm to analyze the data of large-scale users on e-commerce platforms. By clustering user consumption behaviors, this study successfully identified some different types of consumer groups and also revealed some potential trends in consumption upgrades. From the experimental results, DBSCAN has obvious advantages in dealing with scenarios with a lot of noisy data and sparse data distribution. It performs better than the traditional K-means algorithm in terms of clustering accuracy and the level of detail in group division. Therefore, the results of this study actually provide a more accurate method for consumer group division for e-commerce platforms, which can provide data support for subsequent personalized marketing strategies and optimization of product recommendation systems, and is very practical.