Therefore, social media is an excellent resource to automatically discover people who are depressed. It is common for people who suffer from mental health problems too often “implicitly” (and sometimes even “explicitly”) to disclose their feelings and their daily struggles with mental health issues on social media as a way of relief. These sources provide the potential pathway to discover the mental health knowledge for tasks such as diagnosis, medications and claims. Besides sharing their mood and actions, recent studies indicate that many people on social media tend to share or give advice on health-related information.
Individuals and health organizations have shifted away from their traditional interactions, and now meeting online by building online communities for sharing information, seeking and giving the advice to help scale their approach to some extent so that they could cover more affected populations in less time. Moreover, it is very common among people who suffer from depression that they do not visit clinics to ask help from doctors in the early stages of the problem. Diagnosis of depression is usually a difficult task because depression detection needs a thorough and detailed psychological testing by experienced psychiatrists at an early stage and it requires interviews, questionnaires, self-reports or testimony from friends and relatives. One of the common mental health problems is depression that is more dominant than other mental illness conditions worldwide. Sometimes mental illness has been attributed to the mass shooting in the US, which has taken numerous innocent lives. In the United States (US) alone, every year, a significant percentage of the adult population is affected by different mental disorders, which include depression mental illness (6.7%), anorexia and bulimia nervosa (1.6%), and bipolar mental illness (2.6%). Mental illness is a serious issue faced by a large population around the world. MDHAN achieves excellent performance and ensures adequate evidence to explain the prediction. We also show that our model helps improve predictive performance when detecting depression in users who are posting messages publicly on social media. Our experiments show that MDHAN outperforms several popular and robust baseline methods, demonstrating the effectiveness of combining deep learning with multi-aspect features. Our hierarchical attention model is developed in such a way that it can capture patterns that leads to explainable results. Specifically, we encode user posts using two levels of attention mechanisms applied at the tweet-level and word-level, calculate each tweet and words’ importance, and capture semantic sequence features from the user timelines (posts). We have considered user posts augmented with additional features from Twitter. In this work, we propose explainable Multi-Aspect Depression Detection with Hierarchical Attention Network MDHAN, for automatic detection of depressed users on social media and explain the model prediction. For instance, in the task of automatic depression prediction, most machine learning models lead to predictions that are obscure to humans. However, most existing machine learning methods provide no explainability, which is worrying. Model explainability is important for building trust by providing insight into the model prediction. The ability to explain why the model produced results in such a way is an important problem, especially in the medical domain.
0 kommentar(er)
