Feature selection for high-dimensional data: a fast correlation-based filter solution

DMCA

Feature selection for high-dimensional data: a fast correlation-based filter solution (2003)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Lei Yu , Huan Liu

Venue:	In: Proceedings of the 20th International Conferences on Machine Learning
Citations:	274 - 12 self

BibTeX

@INPROCEEDINGS{Yu03featureselection,
    author = {Lei Yu and Huan Liu},
    title = {Feature selection for high-dimensional data: a fast correlation-based filter solution},
    booktitle = {In: Proceedings of the 20th International Conferences on Machine Learning},
    year = {2003}
}

OpenURL

Abstract

Feature selection, as a preprocessing step to machine learning, is effective in reducing di-mensionality, removing irrelevant data, in-creasing learning accuracy, and improving result comprehensibility. However, the re-cent increase of dimensionality of data poses a severe challenge to many existing feature selection methods with respect to efficiency and effectiveness. In this work, we intro-duce a novel concept, predominant correla-tion, and propose a fast filter method which can identify relevant features as well as re-dundancy among relevant features without pairwise correlation analysis. The efficiency and effectiveness of our method is demon-strated through extensive comparisons with other methods using real-world data of high dimensionality. 1.

Keyphrases

feature selection fast correlation-based filter solution high-dimensional data relevant feature pairwise correlation analysis in-creasing learning accuracy preprocessing step feature selection method fast filter method extensive comparison severe challenge irrelevant data real-world data machine learning result comprehensibility predominant correla-tion data pose novel concept re-cent increase high dimensionality

DMCA