Given the current state of the world, because of existing situations around the world, millions of people suffering from mental illnesses feel isolated and unable to receive help in person. Psychological studies have shown that our state of mind can manifest itself in the linguistic features we use to communicate. People have increasingly turned to online platforms to express themselves and seek help with their conditions. Deep learning methods have been commonly used to identify and analyze mental health conditions from various sources of information, including social media. Still, they face challenges, including a lack of reliability and overconfidence in predictions resulting in the poor calibration of the models. To solve these issues, We propose UATTA-EB: Uncertainty-Aware Test-Time Augmented Ensembling of BERTs for producing reliable and well-calibrated predictions to classify six possible types of mental illnesses- None, Depression, Anxiety, Bipolar Disorder, ADHD, and PTSD by analyzing unstructured user data on Reddit.
翻译:鉴于当前世界形势下数百万心理疾病患者感到孤立无助,在现实生活中无法得到帮助。 心理学研究表明,我们的心理状态可能会表现在我们用于交流的语言特征中。人们越来越多地转向在线平台来表达自己并寻求心理健康方面的帮助。 深度学习方法已被普遍用于从各种信息来源,包括社交媒体,识别和分析心理健康状况。 然而,它们面临着挑战,包括缺乏可靠性和过度自信的预测,导致模型的不良校准。为了解决这些问题,我们提出UATTA-EB:基于Bert的测试时间不确定性感知增强Ensemble模型,通过分析 Reddit 上的非结构化用户数据,产生可靠和良好校准的预测,将可能的六种心理疾病 - 无心理疾病,抑郁症, 焦虑症,双向情感障碍,注意力缺陷多动障碍和创伤后应激障碍进行分类。