距離ベースの特徴選択指標

申 吉浩; Adrian Pino Angulo; 久保山 哲二

doi:10.11517/jsaifpai.97.0_08

JSAI Technical Report, SIG-FPAI

Online ISSN : 2436-4584

97th (Mar, 2014)

DOI https://doi.org/10.11517/jsaifpai.97.0_08

Conference information

Host: The Japanese Society for Artificial intelligence

Name : 97th SIG-FPAI

Number : 97

Location : [in Japanese]

Date : March 22, 2014 - March 23, 2014

A Theory of Feature Selection Measures

[in Japanese], [in Japanese], [in Japanese]

Author information

CONFERENCE PROCEEDINGS FREE ACCESS

Pages 08-

Details

Abstract

We present a method to define feature selection measures based on metrics. Also, we show that the well-known Bayesian risk can be derived from some metric and give a new characterization to it.

Content from these authors

Favorites & Alerts

Corresponding author

Conference information

Register with J-STAGE for free!