A Feature Selection Strategy for the Relevance Vector Machine

Armin Shmilovici Leib; David Ben-Shimon

A Feature Selection Strategy for the Relevance Vector Machine

Department of Software and Information Systems Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The Relevance Vector Machine (RVM) is a generalized linear model that can use kernel functions as basis functions. The typical RVM solution is very sparse. We present a strategy for feature ranking and selection via evaluating the influence of the features on the relevance vectors. This requires a single training of
the RVM, thus, it is very efficient. Experiments on a benchmark regression problem provide evidence that it selects high-quality feature sets at a fraction of the costs of classical methods.

Original language	English
Title of host publication	RECENT ADVANCES in KNOWLEDGE ENGINEERING and SYSTEMS SCIENCE
Subtitle of host publication	Proceedings of the 12th International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases (AIKED '13)
Editors	Zengshi Chen, Lopez-Neri Emmanuel
Pages	73-78
State	Published - 2013

Keywords

Feature Selection
Relevance Vector Machine
Machine Learning

Cite this

@inproceedings{fab0693873234f6b93d9c40d022e048e,

title = "A Feature Selection Strategy for the Relevance Vector Machine",

abstract = "The Relevance Vector Machine (RVM) is a generalized linear model that can use kernel functions as basis functions. The typical RVM solution is very sparse. We present a strategy for feature ranking and selection via evaluating the influence of the features on the relevance vectors. This requires a single training ofthe RVM, thus, it is very efficient. Experiments on a benchmark regression problem provide evidence that it selects high-quality feature sets at a fraction of the costs of classical methods. ",

keywords = "Feature Selection, Relevance Vector Machine, Machine Learning",

author = "{Shmilovici Leib}, Armin and David Ben-Shimon",

year = "2013",

language = "English",

pages = "73--78",

editor = "Chen, {Zengshi } and Emmanuel, { Lopez-Neri}",

booktitle = "RECENT ADVANCES in KNOWLEDGE ENGINEERING and SYSTEMS SCIENCE",

}

A Feature Selection Strategy for the Relevance Vector Machine. / Shmilovici Leib, Armin; Ben-Shimon, David.
RECENT ADVANCES in KNOWLEDGE ENGINEERING and SYSTEMS SCIENCE : Proceedings of the 12th International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases (AIKED '13) . ed. / Zengshi Chen; Lopez-Neri Emmanuel. 2013. p. 73-78.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Feature Selection Strategy for the Relevance Vector Machine

AU - Shmilovici Leib, Armin

AU - Ben-Shimon, David

PY - 2013

Y1 - 2013

N2 - The Relevance Vector Machine (RVM) is a generalized linear model that can use kernel functions as basis functions. The typical RVM solution is very sparse. We present a strategy for feature ranking and selection via evaluating the influence of the features on the relevance vectors. This requires a single training ofthe RVM, thus, it is very efficient. Experiments on a benchmark regression problem provide evidence that it selects high-quality feature sets at a fraction of the costs of classical methods.

AB - The Relevance Vector Machine (RVM) is a generalized linear model that can use kernel functions as basis functions. The typical RVM solution is very sparse. We present a strategy for feature ranking and selection via evaluating the influence of the features on the relevance vectors. This requires a single training ofthe RVM, thus, it is very efficient. Experiments on a benchmark regression problem provide evidence that it selects high-quality feature sets at a fraction of the costs of classical methods.

KW - Feature Selection

KW - Relevance Vector Machine

KW - Machine Learning

M3 - Conference contribution

SP - 73

EP - 78

BT - RECENT ADVANCES in KNOWLEDGE ENGINEERING and SYSTEMS SCIENCE

A2 - Chen, Zengshi

A2 - Emmanuel, Lopez-Neri

ER -

A Feature Selection Strategy for the Relevance Vector Machine

Abstract

Keywords

Fingerprint

Cite this