Clustering classifiers learnt from local datasets based on cosine similarity

Kaikai Zhao, Einoshin Suzuki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

In this paper we present a new method to measure the degree of dissimilarity of a pair of linear classifiers. This method is based on the cosine similarity between the normal vectors of the hyperplanes of the linear classifiers. A significant advantage of this method is that it has a good interpretation and requires very little information to exchange among datasets. Evaluations on a synthetic dataset, a dataset from the UCI Machine Learning Repository, and facial expression datasets show that our method outperforms previous methods in terms of the normalized mutual information.

Original languageEnglish
Title of host publicationFoundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings
EditorsFloriana Esposito, Olivier Pivert, Stefano Ferilli, Zbigniew W. Raś, Mohand-Saïd Hacid
PublisherSpringer Verlag
Pages150-159
Number of pages10
ISBN (Print)9783319252513
DOIs
Publication statusPublished - Jan 1 2015
Event22nd International Symposium on Methodologies for Intelligent Systems, ISMIS 2015 - Lyon, France
Duration: Oct 21 2015Oct 23 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9384
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other22nd International Symposium on Methodologies for Intelligent Systems, ISMIS 2015
CountryFrance
CityLyon
Period10/21/1510/23/15

Fingerprint

Classifiers
Classifier
Clustering
Learning systems
Normal vector
Facial Expression
Dissimilarity
Mutual Information
Hyperplane
Repository
Machine Learning
Similarity
Evaluation

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Zhao, K., & Suzuki, E. (2015). Clustering classifiers learnt from local datasets based on cosine similarity. In F. Esposito, O. Pivert, S. Ferilli, Z. W. Raś, & M-S. Hacid (Eds.), Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings (pp. 150-159). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9384). Springer Verlag. https://doi.org/10.1007/978-3-319-25252-0_16

Clustering classifiers learnt from local datasets based on cosine similarity. / Zhao, Kaikai; Suzuki, Einoshin.

Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings. ed. / Floriana Esposito; Olivier Pivert; Stefano Ferilli; Zbigniew W. Raś; Mohand-Saïd Hacid. Springer Verlag, 2015. p. 150-159 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9384).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhao, K & Suzuki, E 2015, Clustering classifiers learnt from local datasets based on cosine similarity. in F Esposito, O Pivert, S Ferilli, ZW Raś & M-S Hacid (eds), Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9384, Springer Verlag, pp. 150-159, 22nd International Symposium on Methodologies for Intelligent Systems, ISMIS 2015, Lyon, France, 10/21/15. https://doi.org/10.1007/978-3-319-25252-0_16
Zhao K, Suzuki E. Clustering classifiers learnt from local datasets based on cosine similarity. In Esposito F, Pivert O, Ferilli S, Raś ZW, Hacid M-S, editors, Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings. Springer Verlag. 2015. p. 150-159. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-25252-0_16
Zhao, Kaikai ; Suzuki, Einoshin. / Clustering classifiers learnt from local datasets based on cosine similarity. Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings. editor / Floriana Esposito ; Olivier Pivert ; Stefano Ferilli ; Zbigniew W. Raś ; Mohand-Saïd Hacid. Springer Verlag, 2015. pp. 150-159 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{29fe69161528473ebc330e619c469e26,
title = "Clustering classifiers learnt from local datasets based on cosine similarity",
abstract = "In this paper we present a new method to measure the degree of dissimilarity of a pair of linear classifiers. This method is based on the cosine similarity between the normal vectors of the hyperplanes of the linear classifiers. A significant advantage of this method is that it has a good interpretation and requires very little information to exchange among datasets. Evaluations on a synthetic dataset, a dataset from the UCI Machine Learning Repository, and facial expression datasets show that our method outperforms previous methods in terms of the normalized mutual information.",
author = "Kaikai Zhao and Einoshin Suzuki",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-3-319-25252-0_16",
language = "English",
isbn = "9783319252513",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "150--159",
editor = "Floriana Esposito and Olivier Pivert and Stefano Ferilli and Raś, {Zbigniew W.} and Mohand-Sa{\"i}d Hacid",
booktitle = "Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings",
address = "Germany",

}

TY - GEN

T1 - Clustering classifiers learnt from local datasets based on cosine similarity

AU - Zhao, Kaikai

AU - Suzuki, Einoshin

PY - 2015/1/1

Y1 - 2015/1/1

N2 - In this paper we present a new method to measure the degree of dissimilarity of a pair of linear classifiers. This method is based on the cosine similarity between the normal vectors of the hyperplanes of the linear classifiers. A significant advantage of this method is that it has a good interpretation and requires very little information to exchange among datasets. Evaluations on a synthetic dataset, a dataset from the UCI Machine Learning Repository, and facial expression datasets show that our method outperforms previous methods in terms of the normalized mutual information.

AB - In this paper we present a new method to measure the degree of dissimilarity of a pair of linear classifiers. This method is based on the cosine similarity between the normal vectors of the hyperplanes of the linear classifiers. A significant advantage of this method is that it has a good interpretation and requires very little information to exchange among datasets. Evaluations on a synthetic dataset, a dataset from the UCI Machine Learning Repository, and facial expression datasets show that our method outperforms previous methods in terms of the normalized mutual information.

UR - http://www.scopus.com/inward/record.url?scp=84952056849&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84952056849&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-25252-0_16

DO - 10.1007/978-3-319-25252-0_16

M3 - Conference contribution

AN - SCOPUS:84952056849

SN - 9783319252513

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

SP - 150

EP - 159

BT - Foundations of Intelligent Systems - 22nd International Symposium, ISMIS 2015, Proceedings

A2 - Esposito, Floriana

A2 - Pivert, Olivier

A2 - Ferilli, Stefano

A2 - Raś, Zbigniew W.

A2 - Hacid, Mohand-Saïd

PB - Springer Verlag

ER -