arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.06326cs.LGcs.AI
收藏

你的二值数据的$\ extit{内在}$维度是什么?——以及如何快速计算它

What is the $\textit{intrinsic}$ dimension of your binary data? -- and how to compute it quickly

University of Hildesheim(希尔德斯海姆大学);University of Kassel(卡塞尔大学);Interdisciplinary Research Center for Information System Design(信息系统设计跨学科研究中心)

Tom Hanika, Tobias Hille

AI总结:

本研究重新审视了二值数据表的归一化相关维度,并将其与最近为几何数据集引入的基于概念的内在维度(ID)进行对比,提出了一种仅计算特定支持度概念的ID快速近似方法,并在469至41271个外在维度的数据集上进行了评估。

摘要:

维度是分析和理解(高维)数据的重要方面。在2006年的ICDM论文中,Tatti等人通过引入归一化相关维度,回答了关于二值数据表(可解释)维度的问题。在本研究中,我们重新审视了他们的结果,并将其与最近为几何数据集引入的基于概念的内在维度(ID)概念进行了对比。为此,我们提出了一种新的ID近似方法,该方法仅计算支持度达到特定值的概念。我们使用Tatti等人提供的所有可用数据集来演示和评估我们的近似方法,这些数据集具有469到41271个外在维度。

英文摘要:

Dimensionality is an important aspect for analyzing and understanding (high-dimensional) data. In their 2006 ICDM paper Tatti et al. answered the question for a (interpretable) dimension of binary data tables by introducing a normalized correlation dimension. In the present work we revisit their results and contrast them with a concept based notion of intrinsic dimension (ID) recently introduced for geometric data sets. To do this, we present a novel approximation for this ID that is based on computing concepts only up to a certain support value. We demonstrate and evaluate our approximation using all available datasets from Tatti et al., which have between 469 and 41271 extrinsic dimensions.

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。

↑