Abstract: This study proposes the implementation of a facial emotion recognition system based on Convolutional Neural Networks to detect emotions in real time, aiming to optimize the workplace environment and enhance organizational productivity. Six deep learning models were evaluated: Standard CNN, AlexNet, VGG16, InceptionV3, ResNet152 and DenseNet201, with DenseNet201 achieving the best performance, delivering an accuracy of 87.7% and recall of 96.3%. The system demonstrated significant improvements in key performance indicators (KPIs), including a 72.59% reduction in data collection time, a 63.4% reduction in diagnosis time, and a 66.59% increase in job satisfaction. These findings highlight the potential of Deep Learning technologies for workplace emotional management, enabling timely interventions and fostering a healthier, more efficient organizational environment.

Cantuarias Valdivia Luis Alberto de Jesús, Gómez Human Javier Junior and Sierra-Liñan Fernando, “Detecting Emotions with Deep Learning Models: Strategies to Optimize the Work Environment and Organizational Productivity” International Journal of Advanced Computer Science and Applications(IJACSA), 16(1), 2025. http://dx.doi.org/10.14569/IJACSA.2025.0160191

@article{Jesús2025,
title = {Detecting Emotions with Deep Learning Models: Strategies to Optimize the Work Environment and Organizational Productivity},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2025.0160191},
url = {http://dx.doi.org/10.14569/IJACSA.2025.0160191},
year = {2025},
publisher = {The Science and Information Organization},
volume = {16},
number = {1},
author = {Cantuarias Valdivia Luis Alberto de Jesús and Gómez Human Javier Junior and Sierra-Liñan Fernando}
}