PERSONET: AN AI-DRIVEN FRAMEWORK FOR PERSONALITY-BASED CUSTOMER SERVICE AGENT MATCHING

Abstract

Because of its many uses, personality categorisation has attracted a lot of attention in the fields of psychology, computational social science, and machine learning (ML). In order to improve customer service experiences by matching clients with appropriate support workers, this article introduces PersoNet, a novel framework designed to determine personality types using the Myers-Briggs Type Indicator (MBTI). PersoNet has an outstanding classification accuracy of over 93.98% and uses a CNN and Bidirectional Long Short-Term Memory (BiLSTM) neural network architecture. The BiLSTM + CNN design successfully captures temporal correlations and semantic complexities in textual data, which contributes to this high degree of accuracy, according to our thorough studies using the MBTI dataset. According to our experimental data, PersoNet can thus precisely choose customer care representatives that fit the personalities of its clients, resulting in a Customer Satisfaction Rate (CSR) of over 97.82%—a noteworthy increase of 20.25% in CSR. These findings position PersoNet as a state-of-the-art personality categorisation tool that significantly improves customer service quality while outperforming current techniques in terms of accuracy and computational economy.