Study of Traditional Approaches and Artificial intelligence-based approach to build Artificial emotional Intelligence Techniques

Abstract

This paper presents a significant advancement in the field of emotion recognition by employing both traditional and machine learning methodologies, tackling the various challenges and limitations that currently exist. Our goal is to perform a comparative analysis of recently implemented machine learning and deep learning algorithms, focusing on those that demonstrate the highest accuracy in emotion detection. The study reviews different feature extraction techniques, classification models, and datasets used for identifying emotions in facial images, speech, and non-verbal cues, providing insights into their characteristics and underlying principles to inform future research directions. Furthermore, we offer an overview of how hybrid classification techniques enhance both accuracy and efficiency in speech emotion recognition. This review aims to contribute to the improvement of automated decision-making services in various customer-centric industries, as well as in patient monitoring within the healthcare sector. The implications of our findings extend to both public and private sectors, including manufacturing industries, highlighting the broad relevance of emotion recognition technologies in enhancing interactions and decision-making processes across diverse applications. By addressing these critical areas, this study seeks to pave the way for more intelligent and responsive systems capable of understanding human emotions effectively.