Predicting Customer Complaint Satisfaction in the Telecommunications Industry and Analyzing Key Influencing Factors Using Artificial Intelligence
Han Yin, Sheng-Bing Wu, Shuping Lu · 2025
The telecommunications industry faces challenges in handling growing volumes of complex customer complaints with traditional manual methods. This study aims to leverage artificial intelligence (AI) and data mining to intelligently analyze complaint data and predict customer satisfaction. Existing approaches lack efficient predictive capabilities. Utilizing the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework on 4,500 authentic complaints, this study identified key satisfaction drivers and developed a fine-tuned BERT model for satisfaction prediction. This data-driven method offers critical insights for enhancing customer experience and demonstrates significant AI potential in proactive management. The developed model achieved high performance, with precision, recall, and F1-scores all exceeding 99%. This research provides telecom companies with actionable tools for improved customer relationship management.