从社交媒体数据检测报告焦虑症状人群中潜在共患 ADHD 的代理指标
Detecting a Proxy for Potential Comorbid ADHD in People Reporting Anxiety Symptoms from Social Media Data
AI总结:
本文利用 Anxiety 与 ADHD subreddit 的发帖迁移行为构建共患代理标签,并微调 RoBERTa 分类器,证明其显著优于关键词方法。
摘要:
我们提出了一项能够阐明焦虑与 ADHD 之间关联的新任务;使用 Transformers 在解决这一无法由基于关键词的分类器完成的任务方面取得进展;并讨论了一种用于可视化我们分类器的方法,以揭示焦虑与 ADHD 表现之间的联系。多达约 50% 的 ADHD 成人可能同时患有焦虑障碍,而约 30% 的焦虑成人可能也患有 ADHD。表现出焦虑的患者可能只接受焦虑治疗,却从未被考虑患有 ADHD,这可能影响治疗。我们展示了如何从社交媒体数据中获得与焦虑共患 ADHD 有关的数据,并表明 Transformers 可用于检测有焦虑症状人群中可能共患 ADHD 的代理指标。我们从焦虑和 ADHD 在线论坛(subreddits)收集数据。我们识别出那些最初开始在 Anxiety subreddit 发帖、后来也开始在 ADHD subreddit 发帖的发布者。我们将这一发布者子集用作代理,代表先表现出焦虑症状、随后意识到自己可能患有 ADHD 的人群。我们微调了一个基于 Transformer 架构的分类器,以区分先开始在 Anxiety subreddit 发帖、随后开始在 ADHD subreddit 发帖的人,与在 Anxiety subreddit 发帖但之后未在 ADHD subreddit 发帖的人。我们表明,Transformer 架构能够取得合理结果:RoBERTa 的正确率为 76%,而最佳基于关键词的模型正确率低于 60%,两者的基准比率均为 50%。
英文摘要:
We present a novel task that can elucidate the connection between anxiety and ADHD; use Transformers to make progress toward solving a task that is not solvable by keyword-based classifiers; and discuss a method for visualization of our classifier illuminating the connection between anxiety and ADHD presentations. Up to approximately 50% of adults with ADHD may also have an anxiety disorder and approximately 30\% of adults with anxiety may also have ADHD. Patients presenting with anxiety may be treated for anxiety without ADHD ever being considered, possibly affecting treatment. We show how data that bears on ADHD that is comorbid with anxiety can be obtained from social media data, and show that Transformers can be used to detect a proxy for possible comorbid ADHD in people with anxiety symptoms. We collected data from anxiety and ADHD online forums (subreddits). We identified posters who first started posting in the Anxiety subreddit and later started posting in the ADHD subreddit as well. We use this subset of the posters as a proxy for people who presented with anxiety symptoms and then became aware that they might have ADHD. We fine-tune a Transformer architecture-based classifier to classify people who started posting in the Anxiety subreddit and then started posting in the ADHD subreddit vs. people who posted in the Anxiety subreddit without later posting in the ADHD subreddit. We show that a Transformer architecture is capable of achieving reasonable results (76% correct for RoBERTa vs. under 60% correct for the best keyword-based model, both with 50% base rate).
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。