AVTO EHTIYOT QISMLARI BOZORIDA MIJOZLARNI SEGMENTATSIYA QILISH UCHUN KLASTERLASH MODELLARI

As a result of the development of digital trading platforms and the automotive industry, the volume of customer data in the auto parts market is increasing significantly. Effective use of this data allows you to identify customer needs, optimize sales strategies, and improve the quality of services. In this research, the issue of customer segmentation in the auto parts market was studied using clustering-based machine learning models. The effectiveness of the K-means, Hierarchical Clustering, DBSCAN, and Gaussian Mixture Model (GMM) algorithms was comparatively analyzed. Experimental studies were conducted based on indicators such as customer purchase frequency, type of spare parts, purchase value, and car brand. The results confirmed that the GMM and DBSCAN algorithms show high accuracy in identifying complex and irregular purchasing behavior.