Predictive Stress Management for Higher Education Faculty Using DCNN and LSTM Models

Teachers in higher education, who are already under a lot of strain from both classroom instruction and administrative duties, need to find ways to alleviate the stress that they experience so that they may be more productive and effective in their roles. The commitment of teachers and the quality of education as a whole are profoundly impacted by their well-being. Deep Convolutional Neural Networks (DCNN) and Long Short-Term Memory (LSTM) models are utilised in this research to present a predictive stress management methodology. Included in the methodology are steps for preparing data, extracting features, and training the model. In the preprocessing phase, extensive changes were made to the text data, and TF-IDF was used to effectively extract features and weight terms. With an average accuracy of 93.50%, the model outperformed standalone CNN and LSTM models thanks to its hybrid CNN-LSTM architecture. According to the results, the suggested method effectively reduces problems caused by stress in educators. This study highlights the model's ability to enhance prediction tasks, which can help institutions implement effective stress management strategies.