iPEERS: A Multi-Layered Expert Recommender System for Enhanced Customer Support

In recent years, there has been a notable surge in digital data and technologies, presenting industrial companies with the potential to enhance their customer service capabilities. While these emerging technologies may pose challenges to established processes, they also represent significant opportunities for growth. Addressing customer inquiries and requests in a prompt and efficient manner is a paramount concern in customer service. Driven by the significant importance and challenges of this issue, we propose an AI-powered multilayered decision support system, called iPEERS, designed to recommend experts who can effectively respond to customer queries. The proposed recommendation process is personalized based on both the customer's profile and the content of their question. Our system integrates various advanced techniques to identify the most suitable expert for each customer inquiry while ensuring an equitable distribution of workload among experts. To illustrate the effectiveness of the approach and demonstrate the functional flow in the proposed solution, we conducted a case study using the StackOverflow dataset spanning from 2008 to 2016. The proposed hybrid recommender system achieved an MAE of 0.723 and an RMSE of 0.889 in predicting the scores given to experts' responses to questions. These metrics demonstrate the system's superior ability to accurately identify the most suitable experts for customer inquiries.