2009 FeatureShapingforLinearSVMClass

• (Forman et al., 2009) ⇒ George Forman, Martin Scholz, and Shyamsundar Rajaram. (2009). “Feature Shaping for Linear SVM Classifiers.” In: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2009). doi:10.1145/1557019.1557057

Subject Headings:

Cited By

• http://scholar.google.com/scholar?q=%22Feature+shaping+for+linear+SVM+classifiers%22+2009
• http://portal.acm.org/citation.cfm?doid=1557019.1557057&preflayout=flat#citedby

Quotes

Author Keywords

Text Classification Machine Learning, Feature Weighting, Feature Scaling, SVM

Abstract

Linear classifiers have been shown to be effective for many discrimination tasks. Irrespective of the learning algorithm itself, the final classifier has a weight to multiply by each feature. This suggests that ideally each input feature should be linearly correlated with the target variable (or anti-correlated), whereas raw features may be highly non-linear. In this paper, we attempt to re-shape each input feature so that it is appropriate to use with a linear weight and to scale the different features in proportion to their predictive value. We demonstrate that this pre-processing is beneficial for linear SVM classifiers on a large benchmark of text classification tasks as well as UCI datasets.

References

,

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2009 FeatureShapingforLinearSVMClass	George Forman Shyamsundar Rajaram Martin Scholz			Feature Shaping for Linear SVM Classifiers		KDD-2009 Proceedings		10.1145/1557019.1557057		2009