摘要
机器人与机器学习的新闻编辑每日新闻-机器学习的新研究是一篇报道的主题。根据来自加纳阿克拉的新闻，由NewsR X记者报道，研究表明，"在行为改变支持系统(BCSSs)中，缺乏关于可靠的一种相关性预测措施的知识。现有的评论主要集中在遵守的自我报告措施上。"我们的新闻记者从加纳大学的研究中得到一句话，“这些措施很容易被高估或低估。本系统综述旨在识别和总结使用机器学习方法预测BCSS依从性的趋势。在2011年1月至2022年8月期间，在Scopus和PubMed电子数据库中进行了系统性文献搜索。初步搜索检索了2182篇Jou Rnal论文，共发现了4类BCSS依从性问题:坚持性认知和行为干预、药物依从性、身体活动依从性、行为依从性利用机器学习技术预测BCSS的实时粘附性正受到越来越多的研究关注。共识别出13种独特的监督学习技术,其中大多数是传统的机器学习技术(如支持向量机)。短期记忆、多层感知和集成学习是目前唯一的高级学习技术。尽管特征选择方法存在着异质性,大多数预测模型具有良好的分类相似性。这表明所使用的特征或预测因子很好地代表了粘附问题。使用机器学习算法预测BCSS用户的粘附行为可以促进粘附行为的强化。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Machine Learning is th e subject of a report. According to news originating from Accra, Ghana, by NewsR x correspondents, research stated, “There is a dearth of knowledge on reliable a dherence prediction measures in behavior change support systems (BCSSs). Existin g reviews have predominately focused on self-reporting measures of adherence.” Our news journalists obtained a quote from the research from the University of G hana, “These measures are susceptible to overestimation or underestimation of ad herence behavior. This systematic review seeks to identify and summarize trends in the use of machine learning approaches to predict adherence to BCSSs. Systema tic literature searches were conducted in the Scopus and PubMed electronic datab ases between January 2011 and August 2022. The initial search retrieved 2182 jou rnal papers, but only 11 of these papers were eligible for this review. A total of 4 categories of adherence problems in BCSSs were identified: adherence to dig ital cognitive and behavioral interventions, medication adherence, physical acti vity adherence, and diet adherence. The use of machine learning techniques for r eal-time adherence prediction in BCSSs is gaining research attention. A total of 13 unique supervised learning techniques were identified and the majority of th em were traditional machine learning techniques (eg, support vector machine). Lo ng short-term memory, multilayer perception, and ensemble learning are currently the only advanced learning techniques. Despite the heterogeneity in the feature selection approaches, most prediction models achieved good classification accur acies. This indicates that the features or predictors used were a good represent ation of the adherence problem. Using machine learning algorithms to predict the adherence behavior of a BCSS user can facilitate the reinforcement of adherence behavior.”
NETL
NSTL