Exploring patterns enriched in a dataset with contrastive principal component analysis.

Exploring patterns enriched in a dataset with contrastive principal component analysis.
复制标题

DOI:
10.1038/s41467-018-04608-8
复制
发表时间:
2018-05-30
影响因子:
16.6
通讯作者:
Zou J
Zou J
中科院分区:
综合性期刊1区
文献类型:
--
作者:
Abid A;Zhang MJ;Bagaria VK;Zou J

文献摘要

参考文献

被引文献

相似文献

高维数据的可视化和探索是跨学科普遍存在的挑战。广泛使用的技术,如主成分分析(PCA),旨在识别一个数据集中的主导趋势。然而,在许多情况下,我们有在不同条件下收集的数据集,例如处理和对照实验,我们对可视化和探索特定于一个数据集的模式感兴趣。本文提出了一种对比主成分分析(CPCA)方法,该方法相对于比较数据识别数据集中丰富的低维结构。在大量的实验中,我们证明了带有背景数据集的CPCA使我们能够可视化PCA和其他标准方法所遗漏的特定于数据集的模式。我们进一步给出了CPCA的几何解释和强大的数学保证。CPCA的实现是公开可用的,并且可以用于当前使用PCA的许多应用中的探索性数据分析。降维和可视化方法缺乏比较多个数据集的原则性方法。在这里,Abid等人。引入对比主成分分析,它识别在一个数据集中相对于另一个数据集中丰富的低维结构,并使数据集特定模式的可视化成为可能。
Visualization and exploration of high-dimensional data is a ubiquitous challenge across disciplines. Widely used techniques such as principal component analysis (PCA) aim to identify dominant trends in one dataset. However, in many settings we have datasets collected under different conditions, e.g., a treatment and a control experiment, and we are interested in visualizing and exploring patterns that are specific to one dataset. This paper proposes a method, contrastive principal component analysis (cPCA), which identifies low-dimensional structures that are enriched in a dataset relative to comparison data. In a wide variety of experiments, we demonstrate that cPCA with a background dataset enables us to visualize dataset-specific patterns missed by PCA and other standard methods. We further provide a geometric interpretation of cPCA and strong mathematical guarantees. An implementation of cPCA is publicly available, and can be used for exploratory data analysis in many applications where PCA is currently used. Dimensionality reduction and visualization methods lack a principled way of comparing multiple datasets. Here, Abid et al. introduce contrastive PCA, which identifies low-dimensional structures enriched in one dataset compared to another and enables visualization of dataset-specific patterns.
DOI: 10.3390/s16030323
发表时间: 2016-03-04
期刊: Sensors (Basel, Switzerland)
影响因子: --
作者:
Chen W;Ma H;Yu D;Zhang H
通讯作者: Zhang H
DOI: 10.1371/journal.pone.0119491
发表时间: 2015
期刊: PloS one
影响因子: 3.7
作者:
Ahmed MM;Dhanasekaran AR;Block A;Tong S;Costa AC;Stasko M;Gardiner KJ
通讯作者: Gardiner KJ
DOI: 10.1126/science.1251688
发表时间: 2014-06-13
期刊: Science (New York, N.Y.)
影响因子: --
作者:
Moreno-Estrada A;Gignoux CR;Fernández-López JC;Zakharia F;Sikora M;Contreras AV;Acuña-Alonzo V;Sandoval K;Eng C;Romero-Hidalgo S;Ortiz-Tello P;Robles V;Kenny EE;Nuño-Arana I;Barquera-Lozano R;Macín-Pérez G;Granados-Arriola J;Huntsman S;Galanter JM;Via M;Ford JG;Chapela R;Rodriguez-Cintron W;Rodríguez-Santana JR;Romieu I;Sienra-Monge JJ;del Rio Navarro B;London SJ;Ruiz-Linares A;Garcia-Herrera R;Estrada K;Hidalgo-Miranda A;Jimenez-Sanchez G;Carnevale A;Soberón X;Canizales-Quinteros S;Rangel-Villalobos H;Silva-Zolezzi I;Burchard EG;Bustamante CD
通讯作者: Bustamante CD
DOI: 10.1073/pnas.0903045106
发表时间: 2009-05-26
影响因子: 11.1
作者:
Silva-Zolezzi, Irma;Hidalgo-Miranda, Alfredo;Jimenez-Sanchez, Gerardo
通讯作者: Jimenez-Sanchez, Gerardo
DOI: 10.1016/j.patcog.2010.12.015
发表时间: 2011-07-01
影响因子: 8
作者:
Barshan, Elnaz;Ghodsi, Ali;Jahromi, Mansoor Zolghadri
通讯作者: Jahromi, Mansoor Zolghadri