TY - JOUR AU - Kabir, Muhammad Khubayeeb AU - Islam, Maisha AU - Kabir, Anika Nahian Binte AU - Haque, Adiba AU - Rhaman, Md Khalilur PY - 2022 DA - 2022/9/28 TI -使用孟加拉语心理健康社交媒体帖子检测抑郁症严重程度:使用自然语言处理技术的研究JO - JMIR Form Res SP - e36118 VL - 6 IS - 9 KW -心理健康论坛KW -自然语言处理KW -严重KW -重度抑郁症KW -深度学习KW -机器学习KW -多类文本分类AB -背景:在书面文本中，有无数的语言线索表明抑郁，自然语言处理(NLP)研究人员已经证明了机器学习和深度学习方法检测这些线索的能力。然而，迄今为止，这些将NLP与孟加拉文学心理健康领域联系起来的方法并不全面。说孟加拉语的人可以更详细地用母语表达情感。目的:我们的目标是通过生成一个新的孟加拉语抑郁帖子语料库来检测抑郁症的严重程度。我们与心理健康专家合作，生成了一个临床合理的标签方案和一个注释语料库，以训练机器学习和深度学习模型。方法:我们使用来自博客和开源平台的孟加拉语文本数据进行了一项研究。我们构建了一个从孟加拉语文献中生成注释语料库和提取文本信息的程序，用于预测分析。我们开发了自己的结构化数据集，并在精神卫生专业人员的帮助下设计了一个临床合理的标签方案，在此过程中遵循了精神疾病诊断和统计手册第五版(DSM-5)。 We used 5 machine learning models for detecting the severity of depression: kernel support vector machine (SVM), random forest, logistic regression K-nearest neighbor (KNN), and complement naive Bayes (NB). For the deep learning approach, we used long short-term memory (LSTM) units and gated recurrent units (GRUs) coupled with convolutional blocks or self-attention layers. Finally, we aimed for enhanced outcomes by using state-of-the-art pretrained language models. Results: The independent recurrent neural network (RNN) models yielded the highest accuracies and weighted F1 scores. GRUs, in particular, produced 81% accuracy. The hybrid architectures could not surpass the RNNs in terms of performance. Kernel SVM with term frequency–inverse document frequency (TF-IDF) embeddings generated 78% accuracy on test data. We used validation and training loss curves to observe and report the performance of our architectures. Overall, the number of available data remained the limitation of our experiment. Conclusions: The findings from our experimental setup indicate that machine learning and deep learning models are fairly capable of assessing the severity of mental health issues from texts. For the future, we suggest more research endeavors to increase the volume of Bengali text data, in particular, so that modern architectures reach improved generalization capability. SN - 2561-326X UR - https://formative.www.mybigtv.com/2022/9/e36118 UR - https://doi.org/10.2196/36118 UR - http://www.ncbi.nlm.nih.gov/pubmed/36169989 DO - 10.2196/36118 ID - info:doi/10.2196/36118 ER -