An Explainable Approach Based on Emotion and Sentiment Features for Detecting People with Mental Disorders on Social Networks

Abstract

Mental disorders are a global problem that widely affects different segments of the population. Diagnosis and treatment are difficult to obtain, as there are not enough specialists on the matter, and mental health is not yet a common topic among the population. The computer science field has proposed some solutions to detect the risk of depression, based on language use and data obtained through social media. These solutions are mainly focused on objective features, such as n-grams and lexicons, which are complicated to be understood by experts in the application area. Hence, in this paper, we propose a contrast pattern-based classifier to detect depression by using a new data representation based only on emotion and sentiment analysis extracted from posts on social media. Our proposed feature representation contains 28 different features, which are more understandable by specialists than other proposed representations. Our feature representation jointly with a contrast pattern-based classifier has obtained better classification results than five other combinations of features and classifiers reported in the literature. Our proposal statistically outperformed the Random Forest, Naive Bayes, and AdaBoost classifiers using the parser-tree, VAD (Valence, Arousal, and Dominance) and Topics, and Bag of Words (BOW) representations. It obtained similar statistical results to the logistic regression models using the Ensemble of BOWs and Handcrafted features representations. In all cases, our proposal was able to provide an explanation close to the language of experts, due to the mined contrast patterns.

Extracting Mental Health Indicators From English and Spanish Social Media: A Machine Learning Approach

“…For future work, we plan to evaluate the performance of PBC4cip [56], a contrast pattern-based classifier that has performed well on tasks such as detecting depressive and xenophobic tweets [57], [58]. Also, compared to popular ML algorithms, it is interpretable, resulting in a set of patterns that provide a deeper understanding of how language is manifested in these mental disorders.…”

Sentiment Informed Sentence BERT-Ensemble Algorithm for Depression Detection

“…They showed KNN was superior with an accuracy of 92%. Authors in Gallegos Salazar et al [42] proposed a contrast pattern-based classifier to detect depression by using a new data representation based only on emotion and sentiment analysis. Similarly, Amanat et al [28] showed RNN-LSTM outperformed SVM, NB, CNN, and decision tress with an accuracy of 99%.…”

Effect of A Selected Corrective Exercise Program on the Function and Proprioception of the Cervical Area in People With Depression and Upper Crossed Syndrome

“…Based on assumptions, the main mechanisms of major depressive disorder include inflammation, disruption, and lack of coordination in hypothalamus function and neurotrophic elements, including neurotrophic factors derived from the brain [4]. Its specific symptoms include loss of energy and interests, low self-confidence, feelings of sadness and guilt, and changes in appetite and sleep patterns, which are associated with differences and similarities in men, women, children, and the elderly, which ultimately reduce the quality of life and the performance of the individuals [5]. One of the important long-term effects of depression is the body's deviation from its normal state.…”