Abstract Millions of people all around the world are affected by depression, a prevalent mental health disease. Effective treatments and assistance depends on the early identification and precise risk assessment of depression. In this study, we provide a unique tuna swarm-optimized attention-based long/short-term memory (TSO-ALSTM) approach based on deep learning (DL) to estimate the risk of depression. The attention mechanism helps the model concentrate on key features, and LSTM is renowned for its capacity to simulate long-term reliance on sequential information. The model's performance is enhanced by using the TSO method to optimize its settings. We make use of a dataset gathered from a Kaggle source to determine the risk of depression. Based on the suggested TSO-ALSTM approach, a number of performance metrics, including AUC, precision, accuracy, recall, and f1-score, are examined in relation to the risk assessment of depression disorder. Results from experiments show how well the suggested TSO-ALSTM model performs in correctly estimating the probability of depression. In regards to accurate forecasting and the value of features, it performs better than other models and conventional machine learning techniques.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|