REVOLUTIONIZING EMOTIONAL INTELLIGENCE ASSESSMENT IN THE MODERN WORKPLACE: INTEGRATING SIGNAL AND IMAGE FUSION WITH CATBOOST

Abstract

In the promptlychangingmodel of the advanced workplace, the significance of emotional intelligence (EI) in employees has gained paramount importance. Employees with higher levels of EI tend to exhibit improved interpersonal relationships, enhanced communication skills, and increased adaptability, all of which contribute to a more productive and harmonious work environment. Addressing the challenge of accurately assessing and enhancing emotional intelligence among employees in the workplace. This study presents a novel method for evaluating and identifying the emotional intelligence of workers by combining signal and image feature fusion with the Catboost Classifier. Convolutional neural networks (CNN), Graph Convolutional Networks (GCN), and self-attention mechanisms are used in conjunction with feature extraction for pictures to automatically extract meaningful patterns for tasks such as image categorization. To improve speech analysis accuracy, adaptive threshold-driven information Gain-Based Feature Extraction (ATI) for speech signals dynamically finds pertinent features. By optimizing correlations across several data sources, Canonical Correlation Analysis (CCA) enables feature fusion and better-informed decision-making. Our Python-based tool achieves an impressive 98% accuracy using FER and TESS datasets, surpassing current techniques.