arXiv (Cornell University) · 2015 · 97 citations · 34 references
ComputingComputer ScienceData ScienceBig Data AnalyticsData MiningMachine LearningData ReductionDimension ReductionDimensionality ReductionOnline AlgorithmsPrincipal Component AnalysisStream ProcessingAccuracyAlgorithm ComparisonDimensionality Reduction For Large DataFunctional Dimension ReductionIncremental Principal Component AnalysisPrincipal ComponentsStreaming Principal Component Analysis
In the current context of data explosion, online techniques that do not\nrequire storing all data in memory are indispensable to routinely perform tasks\nlike principal component analysis (PCA). Recursive algorithms that update the\nPCA with each new observation have been studied in various fields of research\nand found wide applications in industrial monitoring, computer vision,\nastronomy, and latent semantic indexing, among others. This work provides\nguidance for selecting an online PCA algorithm in practice. We present the main\napproaches to online PCA, namely, perturbation techniques, incremental methods,\nand stochastic optimization, and compare their statistical accuracy,\ncomputation time, and memory requirements using artificial and real data.\nExtensions to missing data and to functional data are discussed. All studied\nalgorithms are available in the R package onlinePCA on CRAN.\n
34
Scikit-learn: Machine Learning in Python
Fabián Pedregosa, Gaël Varoquaux, Alexandre Gramfort et al. · arXiv (Cornell University) · 2012 · 63.7K citations · Full text