Cadernos de Linguística e Teoria da Literatura (Universidade Federal de Minas Gerais) · 2018 · 120 citations · 11 references
ComputingComputer ScienceData MiningFeature EngineeringMachine LearningDimension ReductionDimensionality ReductionFeature SelectionFractal GeometryScalable Data MiningScalable Machine LearningAttributesDimensionality Reduction For Large DataFeature Dimensionality ReductionFractal FeaturesHigh-dimensional Data MiningHigh-dimensional Feature SelectionIndexingMultimedia Indexing
Dimensionality curse and dimensionality reduction are two issues that have retained high interest for data mining, machine learning, multimedia indexing, and clustering. We present a fast, scalable algorithm to quickly select the most important attributes (dimensions) for a given set of n-dimensional vectors. In contrast to older methods, our method has the following desirable properties: (a) it does not do rotation of attributes, thus leading to easy interpretation of the resulting attributes; (b) it can spot attributes that have nonlinear correlations; (c) it requires a constant number of passes over the dataset; (d) it gives a good estimate on how many attributes we should keep. The idea is to use the ‘fractal’ dimension of a dataset as a good approximation of its intrinsic dimension, and to drop attributes that do not affect it. We applied our method on real and synthetic datasets, where it gave fast and good results.
11
Matthew Turk, Alex Pentland · Journal of Cognitive Neuroscience · 1991 · 13.8K citations · Full text
Selection of relevant features and examples in machine learning
Avrim Blum, Pat Langley · Artificial Intelligence · 1997 · 3.3K citations
Christos Faloutsos, King-Ip Lin · 1995 · 901 citations · Full text
Intelligent access to digital video: Informedia project
Howard D. Wactlar, Takeo Kanade, Michael A. Smith et al. · Computer · 1996 · 344 citations