Incremental learning of new classes in unbalanced datasets: Learn ++.UDNC

Gregory Ditzler; Michael D. Muhlbaier; Robi Polikar

doi:10.1007/978-3-642-12127-2_4

Incremental learning of new classes in unbalanced datasets: Learn ⁺⁺.UDNC

Gregory Ditzler, Michael D. Muhlbaier, Robi Polikar

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

22 Scopus citations

Abstract

We have previously described an incremental learning algorithm, Learn ⁺⁺.NC, for learning from new datasets that may include new concept classes without accessing previously seen data. We now propose an extension, Learn⁺⁺.UDNC, that allows the algorithm to incrementally learn new concept classes from unbalanced datasets. We describe the algorithm in detail, and provide some experimental results on two separate representative scenarios (on synthetic as well as real world data) along with comparisons to other approaches for incremental and/or unbalanced dataset approaches.

Original language	English (US)
Title of host publication	Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings
Publisher	Springer Verlag
Pages	33-42
Number of pages	10
ISBN (Print)	3642121268, 9783642121265
DOIs	https://doi.org/10.1007/978-3-642-12127-2_4
State	Published - 2010
Externally published	Yes
Event	9th International Workshop on Multiple Classifier Systems, MCS 2010 - Cairo, Egypt Duration: Apr 7 2010 → Apr 9 2010

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5997 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	9th International Workshop on Multiple Classifier Systems, MCS 2010
Country/Territory	Egypt
City	Cairo
Period	4/7/10 → 4/9/10

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-642-12127-2_4

Cite this

Ditzler, G., Muhlbaier, M. D., & Polikar, R. (2010). Incremental learning of new classes in unbalanced datasets: Learn ⁺⁺.UDNC. In Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings (pp. 33-42). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5997 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-642-12127-2_4

Ditzler, Gregory ; Muhlbaier, Michael D. ; Polikar, Robi. / Incremental learning of new classes in unbalanced datasets : Learn ⁺⁺.UDNC. Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings. Springer Verlag, 2010. pp. 33-42 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{c38384828a754e079bbd36a9eb61f827,

title = "Incremental learning of new classes in unbalanced datasets: Learn ++.UDNC",

abstract = "We have previously described an incremental learning algorithm, Learn ++.NC, for learning from new datasets that may include new concept classes without accessing previously seen data. We now propose an extension, Learn++.UDNC, that allows the algorithm to incrementally learn new concept classes from unbalanced datasets. We describe the algorithm in detail, and provide some experimental results on two separate representative scenarios (on synthetic as well as real world data) along with comparisons to other approaches for incremental and/or unbalanced dataset approaches.",

author = "Gregory Ditzler and Muhlbaier, {Michael D.} and Robi Polikar",

year = "2010",

doi = "10.1007/978-3-642-12127-2_4",

language = "English (US)",

isbn = "3642121268",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "33--42",

booktitle = "Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings",

address = "Germany",

note = "9th International Workshop on Multiple Classifier Systems, MCS 2010 ; Conference date: 07-04-2010 Through 09-04-2010",

}

Ditzler, G, Muhlbaier, MD & Polikar, R 2010, Incremental learning of new classes in unbalanced datasets: Learn ⁺⁺.UDNC. in Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5997 LNCS, Springer Verlag, pp. 33-42, 9th International Workshop on Multiple Classifier Systems, MCS 2010, Cairo, Egypt, 4/7/10. https://doi.org/10.1007/978-3-642-12127-2_4

Incremental learning of new classes in unbalanced datasets: Learn ⁺⁺.UDNC. / Ditzler, Gregory; Muhlbaier, Michael D.; Polikar, Robi.
Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings. Springer Verlag, 2010. p. 33-42 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5997 LNCS).

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

TY - GEN

T1 - Incremental learning of new classes in unbalanced datasets

T2 - 9th International Workshop on Multiple Classifier Systems, MCS 2010

AU - Ditzler, Gregory

AU - Muhlbaier, Michael D.

AU - Polikar, Robi

PY - 2010

Y1 - 2010

N2 - We have previously described an incremental learning algorithm, Learn ++.NC, for learning from new datasets that may include new concept classes without accessing previously seen data. We now propose an extension, Learn++.UDNC, that allows the algorithm to incrementally learn new concept classes from unbalanced datasets. We describe the algorithm in detail, and provide some experimental results on two separate representative scenarios (on synthetic as well as real world data) along with comparisons to other approaches for incremental and/or unbalanced dataset approaches.

AB - We have previously described an incremental learning algorithm, Learn ++.NC, for learning from new datasets that may include new concept classes without accessing previously seen data. We now propose an extension, Learn++.UDNC, that allows the algorithm to incrementally learn new concept classes from unbalanced datasets. We describe the algorithm in detail, and provide some experimental results on two separate representative scenarios (on synthetic as well as real world data) along with comparisons to other approaches for incremental and/or unbalanced dataset approaches.

UR - http://www.scopus.com/inward/record.url?scp=77952047336&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952047336&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-12127-2_4

DO - 10.1007/978-3-642-12127-2_4

M3 - Conference contribution

AN - SCOPUS:77952047336

SN - 3642121268

SN - 9783642121265

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 33

EP - 42

BT - Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings

PB - Springer Verlag

Y2 - 7 April 2010 through 9 April 2010

ER -

Ditzler G, Muhlbaier MD, Polikar R. Incremental learning of new classes in unbalanced datasets: Learn ⁺⁺.UDNC. In Multiple Classifier Systems - 9th International Workshop, MCS 2010, Proceedings. Springer Verlag. 2010. p. 33-42. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-12127-2_4