为了解决传统方法识别自杀风险的局限性,研究普通人群中自杀行为的研究人员呼吁将重点从先验风险因素转变为基于机器学习的算法。14具有大型数据集,机器学习方法提供了几种优势,包括使用可以通过建模变量之间的复杂关联来促进临床意义和概括性的技术,以及自动学习优化预测算法的方法而不是依赖研究人员的先验假设的方法。14,15个机器学习方法在预测异质且罕见发生的结果(即自杀思想和行为)方面可能是有利的,这些结果通常与阶级失衡有关。具体来说,可以与传统的机器学习程序结合实施采样程序,包括缩写,以预测不平衡的结果。Using a combination of surveys, administrative data, and electronic health records, machine learning has shown promise in identifying short- and long-term risk for suicidal behaviors (in non-TBI populations) with greater accuracy than traditional approaches, up to several years before they occur and across a variety of settings, including the US Army, 16,17 Veterans Health Administration, 18 independent health care systems, 15,19,20 and population-wide studies.21,22
主要关键词