Relevance as a metric for evaluating machine learning algorithms

A. Kota Gopalakrishna; T. Ozcelebi; A. Liotta; J.J. Lukkien

doi:10.1007/978-3-642-39712-7_15

Relevance as a metric for evaluating machine learning algorithms

A. Kota Gopalakrishna, T. Ozcelebi, A. Liotta, J.J. Lukkien

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

11 Citations (Scopus)

1 Downloads (Pure)

Abstract

In machine learning, the choice of a learning algorithm that is suitable for the application domain is critical. The performance metric used to compare different algorithms must also reflect the concerns of users in the application domain under consideration. In this paper, we propose a novel probability-based performance metric called Relevance Score for evaluating supervised learning algorithms. We evaluate the proposed metric through empirical analysis on a dataset gathered from an intelligent lighting pilot installation. In comparison to the commonly used Classification Accuracy metric, the Relevance Score proves to be more appropriate for a certain class of applications.

Original language	English
Title of host publication	Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings)
Editors	P. Perner
Place of Publication	Berlin
Publisher	Springer
Pages	195-208
ISBN (Print)	978-3-642-39711-0
DOIs	https://doi.org/10.1007/978-3-642-39712-7_15
Publication status	Published - 2013
Event	9th International Conmference on Machine Learning and Data Mining in Pattern Recognition (MLDM 2013) - New York, United States Duration: 19 Jun 2013 → 25 Jun 2013 Conference number: 9

Publication series

Name	Lecture Notes in Computer Science
Volume	7988
ISSN (Print)	0302-9743

Conference

Conference	9th International Conmference on Machine Learning and Data Mining in Pattern Recognition (MLDM 2013)
Abbreviated title	MLDM 2013
Country/Territory	United States
City	New York
Period	19/06/13 → 25/06/13

Access to Document

10.1007/978-3-642-39712-7_15

Cite this

Kota Gopalakrishna, A., Ozcelebi, T., Liotta, A., & Lukkien, J. J. (2013). Relevance as a metric for evaluating machine learning algorithms. In P. Perner (Ed.), Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings) (pp. 195-208). (Lecture Notes in Computer Science; Vol. 7988). Springer. https://doi.org/10.1007/978-3-642-39712-7_15

@inproceedings{87847076bead4e8e975b4719dde0ca68,

title = "Relevance as a metric for evaluating machine learning algorithms",

abstract = "In machine learning, the choice of a learning algorithm that is suitable for the application domain is critical. The performance metric used to compare different algorithms must also reflect the concerns of users in the application domain under consideration. In this paper, we propose a novel probability-based performance metric called Relevance Score for evaluating supervised learning algorithms. We evaluate the proposed metric through empirical analysis on a dataset gathered from an intelligent lighting pilot installation. In comparison to the commonly used Classification Accuracy metric, the Relevance Score proves to be more appropriate for a certain class of applications.",

author = "{Kota Gopalakrishna}, A. and T. Ozcelebi and A. Liotta and J.J. Lukkien",

year = "2013",

doi = "10.1007/978-3-642-39712-7_15",

language = "English",

isbn = "978-3-642-39711-0",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "195--208",

editor = "P. Perner",

booktitle = "Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings)",

address = "Germany",

note = "9th International Conmference on Machine Learning and Data Mining in Pattern Recognition (MLDM 2013), MLDM 2013 ; Conference date: 19-06-2013 Through 25-06-2013",

}

Kota Gopalakrishna, A, Ozcelebi, T, Liotta, A & Lukkien, JJ 2013, Relevance as a metric for evaluating machine learning algorithms. in P Perner (ed.), Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings). Lecture Notes in Computer Science, vol. 7988, Springer, Berlin, pp. 195-208, 9th International Conmference on Machine Learning and Data Mining in Pattern Recognition (MLDM 2013), New York, New York, United States, 19/06/13. https://doi.org/10.1007/978-3-642-39712-7_15

Relevance as a metric for evaluating machine learning algorithms. / Kota Gopalakrishna, A.; Ozcelebi, T.; Liotta, A. et al.

Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings). ed. / P. Perner. Berlin : Springer, 2013. p. 195-208 (Lecture Notes in Computer Science; Vol. 7988).

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

TY - GEN

T1 - Relevance as a metric for evaluating machine learning algorithms

AU - Kota Gopalakrishna, A.

AU - Ozcelebi, T.

AU - Liotta, A.

AU - Lukkien, J.J.

N1 - Conference code: 9

PY - 2013

Y1 - 2013

N2 - In machine learning, the choice of a learning algorithm that is suitable for the application domain is critical. The performance metric used to compare different algorithms must also reflect the concerns of users in the application domain under consideration. In this paper, we propose a novel probability-based performance metric called Relevance Score for evaluating supervised learning algorithms. We evaluate the proposed metric through empirical analysis on a dataset gathered from an intelligent lighting pilot installation. In comparison to the commonly used Classification Accuracy metric, the Relevance Score proves to be more appropriate for a certain class of applications.

AB - In machine learning, the choice of a learning algorithm that is suitable for the application domain is critical. The performance metric used to compare different algorithms must also reflect the concerns of users in the application domain under consideration. In this paper, we propose a novel probability-based performance metric called Relevance Score for evaluating supervised learning algorithms. We evaluate the proposed metric through empirical analysis on a dataset gathered from an intelligent lighting pilot installation. In comparison to the commonly used Classification Accuracy metric, the Relevance Score proves to be more appropriate for a certain class of applications.

U2 - 10.1007/978-3-642-39712-7_15

DO - 10.1007/978-3-642-39712-7_15

M3 - Conference contribution

SN - 978-3-642-39711-0

T3 - Lecture Notes in Computer Science

SP - 195

EP - 208

BT - Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings)

A2 - Perner, P.

PB - Springer

CY - Berlin

T2 - 9th International Conmference on Machine Learning and Data Mining in Pattern Recognition (MLDM 2013)

Y2 - 19 June 2013 through 25 June 2013

ER -

Kota Gopalakrishna A, Ozcelebi T, Liotta A, Lukkien JJ. Relevance as a metric for evaluating machine learning algorithms. In Perner P, editor, Machine Learning and Data Mining (9th International Conference, MLDM 2013, New York NY, USA, July 19-25, 2013. Proceedings). Berlin: Springer. 2013. p. 195-208. (Lecture Notes in Computer Science). doi: 10.1007/978-3-642-39712-7_15

Relevance as a metric for evaluating machine learning algorithms

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this