Multi-domain Data Association Analysis: Research on Precise Customer Classification Based on LLM and GMM Models

This study delves into the application of multi-domain data association analysis in precise customer classification, proposing an innovative method based on the Large Language Model (LLM) and the Gaussian Mixture Model (GMM). With the advent of the big data era, the massive customer data that enterprises face exhibit a high degree of complexity and diversity, making traditional customer group analysis methods no longer sufficient to meet the needs of precise analysis. The study first introduces the advantages of the LLM model in text similarity comparison and the flexibility of the GMM model in handling the clustering of multi-dimensional data. It then elaborates on the construction process of the customer segmentation model that combines LLM and GMM models, including key steps such as the selection and feature extraction of the LLM model, and the construction and training of the GMM model. Through empirical analysis, the effectiveness of the model is verified using a banking dataset, and a comparison is made with the traditional Kmeans algorithm. The results show that the new model demonstrates significant advantages in clustering effects and data processing capabilities. This research not only enriches the theory and methods in the field of customer segmentation but also provides a scientific basis and practical tools for enterprises to formulate differentiated service policies and precise marketing strategies.