Call Center Customer Sentiment Analysis Using ML and NLP

Abstract

In the contemporary digital era, call centers have significantly incorporated automation through callbots, but they often lack an essential aspect of customer service: empathy. This paper explores the integration of sentiment analysis into call center operations to introduce an emotional dimension to callbot interactions. Utilizing natural language processing and machine learning, the paper examines both text-based and signal-based sentiment analysis approaches. The proposed sentiment analysis architecture encompasses data input and output interfaces, preprocessing, sentiment analysis, and a decision manager module that can, for example, escalate calls to a human agent based on sentiment analysis results. The preprocessing steps, both for text and signal analysis, are outlined in detail, along with a review of relevant sentiment analysis algorithms.Through comprehensive experiments, the study demonstrates that the integration of sentiment analysis yields promising outcomes. In text-based analysis, SVM and LSTM models consistently performed well, achieving accuracy scores of 74% and 72%, respectively. For voice-based analysis, the MLP model exhibited the highest accuracy of 0.72 when using mel spectrogram features, while the RF model outperformed others with an accuracy of 0.78 using MFCC features. These results showcase the potential of sentiment analysis in humanizing callbot interactions, thereby enhancing customer satisfaction and service efficiency.