摘要
由一名新闻记者兼机器人与机器学习的工作人员新闻编辑每日新闻-智能系统的最新数据在一份新的报告中呈现。根据匈牙利布达佩斯的新闻报道，B y NewsRx记者，研究表明，"班级失衡学习在训练数据集在特定班级中表现出不成比例的样本的各个领域具有挑战性。"新闻记者引用了Eotvos Lorand Univers(ELTE)的一篇研究文章:“重采样方法已经被用来调整类别分布,但它们对于小的分离少数子集往往有局限性。”针对这些局限性，提出了一种基于聚类的自适应过采样方法。AROSS利用共相关系数和贝叶斯信息准则优化聚集聚类算法识别少数群体的代表性区域，采用增量K近邻策略获得安全区域和半安全区域，采用截断超球面高斯分布进行过采样。根据新闻编辑的说法，该研究得出结论：“在70个二进制数据集上的实验评估证明了AROSS在改善班级失衡学习表现方面的有效性，使其成为缓解C类失衡挑战的有希望的解决方案，特别是对小的分离少数子集。”
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Fresh data on intelligent systems are presented in a new report. According to news reporting from Budapest, Hungary, b y NewsRx journalists, research stated, “Class imbalance learning is challenging in various domains where training datasets exhibit disproportionate samples in a specific class.” The news reporters obtained a quote from the research from Eotvos Lorand Univers ity (ELTE): “Resampling methods have been used to adjust the class distribution, but they often have limitations for small disjunct minority subsets. This paper introduces AROSS, an adaptive cluster-based oversampling approach that addresse s these limitations. AROSS utilizes an optimized agglomerative clustering algori thm with the Cophenetic Correlation Coefficient and the Bayesian Information Cri terion to identify representative areas of the minority class. Safe and half-saf e areas are obtained using an incremental k-Nearest Neighbor strategy, and overs ampling is performed with a truncated hyperspherical Gaussian distribution.” According to the news editors, the research concluded: “Experimental evaluations on 70 binary datasets demonstrate the effectiveness of AROSS in improving class imbalance learning performance, making it a promising solution for mitigating c lass imbalance challenges, especially for small disjunct minority subsets.”
NETL
NSTL