Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised

Research output: Contribution to conferencePaperResearchpeer-review

Standard

Learning Outlier Ensembles : The Best of Both Worlds–Supervised and Unsupervised. / Micenková, Barbora; McWilliams, Brian ; Assent, Ira.

2014. Paper presented at ACM SIGKDD 2014 Workshop ODD2 , New York, United States.

Research output: Contribution to conferencePaperResearchpeer-review

Harvard

Micenková, B, McWilliams, B & Assent, I 2014, 'Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised', Paper presented at ACM SIGKDD 2014 Workshop ODD2 , New York, United States, 24/08/2014 - 24/08/2014. <http://outlier-analytics.org/odd14kdd/odd-2014-proceedings.pdf>

APA

Micenková, B., McWilliams, B., & Assent, I. (2014). Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised. Paper presented at ACM SIGKDD 2014 Workshop ODD2 , New York, United States. http://outlier-analytics.org/odd14kdd/odd-2014-proceedings.pdf

CBE

Micenková B, McWilliams B, Assent I. 2014. Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised. Paper presented at ACM SIGKDD 2014 Workshop ODD2 , New York, United States.

MLA

Micenková, Barbora, Brian McWilliams, and Ira Assent Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised. ACM SIGKDD 2014 Workshop ODD2 , 24 Aug 2014, New York, United States, Paper, 2014. 4 p.

Vancouver

Micenková B, McWilliams B, Assent I. Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised. 2014. Paper presented at ACM SIGKDD 2014 Workshop ODD2 , New York, United States.

Author

Micenková, Barbora ; McWilliams, Brian ; Assent, Ira. / Learning Outlier Ensembles : The Best of Both Worlds–Supervised and Unsupervised. Paper presented at ACM SIGKDD 2014 Workshop ODD2 , New York, United States.4 p.

Bibtex

@conference{a4dfe81e04d14cdbb3dc060eb06b7f48,
title = "Learning Outlier Ensembles: The Best of Both Worlds–Supervised and Unsupervised",
abstract = "Years of research in unsupervised outlier detection have produced numerous algorithms to score data according to their exceptionality. wever, the nature of outliers heavily depends on the application context and different algorithms are sensitive to outliers of different nature. This makes it very difficult to assess suitability of a particular algorithm without a priori knowledge. On the other hand, in many applications, some examples of outliers exist or can be obtainedin addition to the vast amount of unlabeled data. Unfortunately, this extra knowledge cannot be simply incorporated into the existing unsupervised algorithms.In this paper, we show how to use powerful machine learning approaches to combine labeled examples together with arbitrary unsupervised outlier scoring algorithms. We aim to get the best out of the two worlds—supervised and unsupervised. Our approach is also a viable solution to the recent problem of outlier ensemble selection.",
keywords = "Outlier detection, utlier ensembles, semi-supervised ou tlier detection",
author = "Barbora Micenkov{\'a} and Brian McWilliams and Ira Assent",
year = "2014",
language = "English",
note = "null ; Conference date: 24-08-2014 Through 24-08-2014",

}

RIS

TY - CONF

T1 - Learning Outlier Ensembles

AU - Micenková, Barbora

AU - McWilliams, Brian

AU - Assent, Ira

N1 - Conference code: 2

PY - 2014

Y1 - 2014

N2 - Years of research in unsupervised outlier detection have produced numerous algorithms to score data according to their exceptionality. wever, the nature of outliers heavily depends on the application context and different algorithms are sensitive to outliers of different nature. This makes it very difficult to assess suitability of a particular algorithm without a priori knowledge. On the other hand, in many applications, some examples of outliers exist or can be obtainedin addition to the vast amount of unlabeled data. Unfortunately, this extra knowledge cannot be simply incorporated into the existing unsupervised algorithms.In this paper, we show how to use powerful machine learning approaches to combine labeled examples together with arbitrary unsupervised outlier scoring algorithms. We aim to get the best out of the two worlds—supervised and unsupervised. Our approach is also a viable solution to the recent problem of outlier ensemble selection.

AB - Years of research in unsupervised outlier detection have produced numerous algorithms to score data according to their exceptionality. wever, the nature of outliers heavily depends on the application context and different algorithms are sensitive to outliers of different nature. This makes it very difficult to assess suitability of a particular algorithm without a priori knowledge. On the other hand, in many applications, some examples of outliers exist or can be obtainedin addition to the vast amount of unlabeled data. Unfortunately, this extra knowledge cannot be simply incorporated into the existing unsupervised algorithms.In this paper, we show how to use powerful machine learning approaches to combine labeled examples together with arbitrary unsupervised outlier scoring algorithms. We aim to get the best out of the two worlds—supervised and unsupervised. Our approach is also a viable solution to the recent problem of outlier ensemble selection.

KW - Outlier detection

KW - utlier ensembles

KW - semi-supervised ou tlier detection

M3 - Paper

Y2 - 24 August 2014 through 24 August 2014

ER -