Abstract—Recognition of facial expressions in real-time has established itself as a prominent area of research. This field explores machine learning techniques to analyze expressions from text, voice, and facial features. In this context, we present Dainattor, derived from the Latin term “Inordinatio-prediktor”, which means predictor of emotional disorders. This web system specializes in identifying the seven universal facial expressions: anger, disgust, fear, happiness, neutrality, sadness, and surprise. It uses advanced machine learning techniques to monitor these expressions and predict emotional disturbances, based on a history of facial data. The research was divided into two main phases: in the first phase, a systematic mapping study was conducted to identify relevant research that would guide the development of Dainattor. In the second phase, the Extreme Programming methodology was implemented to design the system, incorporating the FER13 dataset to train a convolutional neural network. This model achieved an accuracy of 86.28% after 50 epochs. The polynomial regression technique was also used to predict emotional disorders. The usability of Dainattor was evaluated through Tree Testing with ten users, confirming its effectiveness in recognizing facial expressions and predicting emotional disorders in a satisfactory manner.
 
Keywords—machine learning, facial recognition, facial expression detection, prediction of emotional disorders, convolutional neural network


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).