Stress Detection During Social Interactions with Natural Language Processing and Machine Learning

The field of Natural Language Processing (NLP) is witnessing a growing recognition of the significance of investigating stress-related phenomena. This research study proposes a new framework to address the diagnostic challenges associated with mental health conditions such as anxiety, depression, and stress, which can worsen if left unaddressed. Within this research study, social media postings and comments are used to enhance sentiment and emotion analysis for assessing individuals' stress levels. Specifically, text analysis of social media data, including Twitter tweets, serves as a pivotal application. The proposed system is implemented by employing a combination of Machine Learning (ML), Natural Language Processing (NLP), and Deep Learning (DL) methodologies. The proposed system utilizes Convolutional Neural Network (CNN), BERT transfer learning techniques, as well as Machine Learning (ML) algorithms such as Naive Bayes, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). This study aims to predict five distinct emotional categories: “Happy, ” “Angry, ” “Depression, ” “Sad, ” and “Fear.” Remarkably, the proposed system has achieved an accuracy and F1 score exceeding 93%, outperforming the performance of other existing human emotion recognition systems.