Optimizing call center agent efficiency through deep learning-based classifications using SMFCCAE

Abstract

Call centers are vital to business operations worldwide, acting as the primary interface between companies and their customers. They handle customer inquiries, manage complaints, and facilitate telephonic sales, making them essential to customer service. However, ensuring quality in the call center industry remains challenging, primarily due to the heavy reliance on call center representatives (CSRs) who manage high volumes of calls. Traditional methods of evaluating CSR performance often rely on manual assessments of small call samples, which can be time-consuming and limited in scope. With the advancement of deep learning techniques (DLTs), there is an opportunity to more accurately assess CSR performance. This study introduces the selecting minimal features for call center agents efficiency (SMFCCE) approach, which optimizes feature selection from CSR data to enhance classification accuracy and speed. The proposed method achieves approximately 85% accuracy, offering valuable insights and recommendations for improving overall call center operations.