Measuring similarity between matrix objects for pattern recognition

Hyunsoek Choi; Hyeyoung Park

doi:10.1145/2814940.2814967

Measuring similarity between matrix objects for pattern recognition

Hyunsoek Choi, Hyeyoung Park

Kyungpook National University

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

1 Scopus citations

Abstract

In order to make machines able to recognize various patterns, it is important to define an appropriate function for measuring similarities between different objects. Conventional similarity measures are devised mainly for 1D vector data, which may lead to loss of information of 2D matrix data. We cast the calculation of similarity between two matrices as a neural network problem, and design the architecture for learning a similarity measure. We provide experiments on real 2D matrix data in the face recognition and gesture recognition, where we show that the learning of a similarity measure leads to improvements in the performance of the recognition problem. Also we compare the performance of the proposed measure with conventional distance measures for 2D matrix data.

Original language	English
Title of host publication	HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction
Publisher	Association for Computing Machinery, Inc
Pages	175-177
Number of pages	3
ISBN (Electronic)	9781450335270
DOIs	https://doi.org/10.1145/2814940.2814967
State	Published - 21 Oct 2015
Event	3rd International Conference on Human-Agent Interaction, HAI 2015 - Daegu, Korea, Republic of Duration: 21 Oct 2015 → 24 Oct 2015

Publication series

Name	HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction

Conference

Conference	3rd International Conference on Human-Agent Interaction, HAI 2015
Country/Territory	Korea, Republic of
City	Daegu
Period	21/10/15 → 24/10/15

Keywords

Matrix data
Neural networks
Similarity measure

Access to Document

10.1145/2814940.2814967

Cite this

Choi, H., & Park, H. (2015). Measuring similarity between matrix objects for pattern recognition. In HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction (pp. 175-177). (HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction). Association for Computing Machinery, Inc. https://doi.org/10.1145/2814940.2814967

@inproceedings{11976b53fbc04b86a11a650d5829bdfb,

title = "Measuring similarity between matrix objects for pattern recognition",

abstract = "In order to make machines able to recognize various patterns, it is important to define an appropriate function for measuring similarities between different objects. Conventional similarity measures are devised mainly for 1D vector data, which may lead to loss of information of 2D matrix data. We cast the calculation of similarity between two matrices as a neural network problem, and design the architecture for learning a similarity measure. We provide experiments on real 2D matrix data in the face recognition and gesture recognition, where we show that the learning of a similarity measure leads to improvements in the performance of the recognition problem. Also we compare the performance of the proposed measure with conventional distance measures for 2D matrix data.",

keywords = "Matrix data, Neural networks, Similarity measure",

author = "Hyunsoek Choi and Hyeyoung Park",

note = "Publisher Copyright: {\textcopyright} 2015 ACM.; 3rd International Conference on Human-Agent Interaction, HAI 2015 ; Conference date: 21-10-2015 Through 24-10-2015",

year = "2015",

month = oct,

day = "21",

doi = "10.1145/2814940.2814967",

language = "English",

series = "HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction",

publisher = "Association for Computing Machinery, Inc",

pages = "175--177",

booktitle = "HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction",

}

Choi, H & Park, H 2015, Measuring similarity between matrix objects for pattern recognition. in HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction. HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction, Association for Computing Machinery, Inc, pp. 175-177, 3rd International Conference on Human-Agent Interaction, HAI 2015, Daegu, Korea, Republic of, 21/10/15. https://doi.org/10.1145/2814940.2814967

Measuring similarity between matrix objects for pattern recognition. / Choi, Hyunsoek; Park, Hyeyoung.
HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction. Association for Computing Machinery, Inc, 2015. p. 175-177 (HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction).

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

TY - GEN

T1 - Measuring similarity between matrix objects for pattern recognition

AU - Choi, Hyunsoek

AU - Park, Hyeyoung

PY - 2015/10/21

Y1 - 2015/10/21

N2 - In order to make machines able to recognize various patterns, it is important to define an appropriate function for measuring similarities between different objects. Conventional similarity measures are devised mainly for 1D vector data, which may lead to loss of information of 2D matrix data. We cast the calculation of similarity between two matrices as a neural network problem, and design the architecture for learning a similarity measure. We provide experiments on real 2D matrix data in the face recognition and gesture recognition, where we show that the learning of a similarity measure leads to improvements in the performance of the recognition problem. Also we compare the performance of the proposed measure with conventional distance measures for 2D matrix data.

AB - In order to make machines able to recognize various patterns, it is important to define an appropriate function for measuring similarities between different objects. Conventional similarity measures are devised mainly for 1D vector data, which may lead to loss of information of 2D matrix data. We cast the calculation of similarity between two matrices as a neural network problem, and design the architecture for learning a similarity measure. We provide experiments on real 2D matrix data in the face recognition and gesture recognition, where we show that the learning of a similarity measure leads to improvements in the performance of the recognition problem. Also we compare the performance of the proposed measure with conventional distance measures for 2D matrix data.

KW - Matrix data

KW - Neural networks

KW - Similarity measure

UR - http://www.scopus.com/inward/record.url?scp=84962885917&partnerID=8YFLogxK

U2 - 10.1145/2814940.2814967

DO - 10.1145/2814940.2814967

M3 - Conference contribution

AN - SCOPUS:84962885917

T3 - HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction

SP - 175

EP - 177

BT - HAI 2015 - Proceedings of the 3rd International Conference on Human-Agent Interaction

PB - Association for Computing Machinery, Inc

T2 - 3rd International Conference on Human-Agent Interaction, HAI 2015

Y2 - 21 October 2015 through 24 October 2015

ER -

Measuring similarity between matrix objects for pattern recognition

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this