Subgroup Agrawal Drift Dataset
收藏资源简介:
Subgroup Agrawal Drift Dataset是由都灵理工大学和加州大学圣克鲁兹分校的研究团队基于Agrawal生成器创建的合成数据集,旨在模拟数据流中的局部漂移现象。数据集包含10,000条记录,具有六种数值属性和三种分类属性,用于二元分类任务。数据集的创建过程中,通过随机选择子群体并引入噪声来模拟局部漂移。该数据集主要用于评估和改进现有的漂移检测技术,特别是在局部漂移检测方面的性能。
The Subgroup Agrawal Drift Dataset is a synthetic dataset developed by research teams from Politecnico di Torino and the University of California, Santa Cruz, based on the Agrawal generator. It aims to simulate local drift phenomena in data streams. This dataset contains 10,000 records, with six numerical attributes and three categorical attributes, and is designed for binary classification tasks. During its creation, local drift is simulated by randomly selecting subgroups and introducing noise. This dataset is primarily used to evaluate and improve existing drift detection technologies, particularly their performance in local drift detection.

- 1A Synthetic Benchmark to Explore Limitations of Localized Drift Detections都灵理工大学, 加州大学圣克鲁兹分校 · 2024年



