On the noise resilience of ranking measures

Daniel Berrar

研究成果: Conference contribution

2 被引用数 (Scopus)

抄録

Performance measures play a pivotal role in the evaluation and selection of machine learning models for a wide range of applications. Using both synthetic and real-world data sets, we investigated the resilience to noise of various ranking measures. Our experiments revealed that the area under the ROC curve (AUC) and a related measure, the truncated average Kolmogorov-Smirnov statistic (taKS), can reliably discriminate between models with truly different performance under various types and levels of noise. With increasing class skew, however, the H-measure and estimators of the area under the precision-recall curve become preferable measures. Because of its simple graphical interpretation and robustness, the lower trapezoid estimator of the area under the precision-recall curve is recommended for highly imbalanced data sets.

本文言語English
ホスト出版物のタイトルNeural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
編集者Seiichi Ozawa, Kazushi Ikeda, Derong Liu, Akira Hirose, Kenji Doya, Minho Lee
出版社Springer Verlag
ページ47-55
ページ数9
ISBN(印刷版)9783319466712
DOI
出版ステータスPublished - 2016
イベント23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, Japan
継続期間: 2016 10月 162016 10月 21

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
9948 LNCS
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Other

Other23rd International Conference on Neural Information Processing, ICONIP 2016
国/地域Japan
CityKyoto
Period16/10/1616/10/21

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータサイエンス一般

フィンガープリント

「On the noise resilience of ranking measures」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル