Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification

Md Hafizur Rahman; Nizar Bouguila

doi:10.1109/SMC42975.2020.9283063

Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification

Md Hafizur Rahman, Nizar Bouguila

Bloomberg School of Public Health

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

Abstract

In this paper, we propose a distribution based feature mapping technique to improve the baseline accuracy of SVM kernels in different computer vision tasks. The proposed technique is based on learning parameters from the data and use that parameters to make inference from new data. The learned parameters can be thought of as prior knowledge about the data representation. Utilizing such prior knowledge about the data distribution increases the discriminative power of the classifier. Our proposed feature mapping technique is based on inverted Dirichlet, generalized inverted Dirichlet and inverted Beta Liouville distributions. These distributions are efficient in modelling semi-bounded data which are prevalent in computer vision problems. Our experimental results demonstrate the effectiveness of the proposed method in texture recognition, natural scene recognition and human action recognition in videos.

Original language	English (US)
Title of host publication	2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3588-3593
Number of pages	6
Volume	2020-October
ISBN (Electronic)	9781728185262
DOIs	https://doi.org/10.1109/SMC42975.2020.9283063
State	Published - Oct 11 2020
Event	2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020 - Toronto, Canada Duration: Oct 11 2020 → Oct 14 2020

Conference

Conference	2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020
Country	Canada
City	Toronto
Period	10/11/20 → 10/14/20

Keywords

Feature Mapping
Generalized Inverted Dirichlet Distribution
Inverted Beta-Liouville Distribution
Inverted Dirichlet Distribution
SVM

ASJC Scopus subject areas

Software
Control and Systems Engineering
Human-Computer Interaction
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/SMC42975.2020.9283063

Fingerprint Dive into the research topics of 'Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification'. Together they form a unique fingerprint.

Cite this

Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification. / Rahman, Md Hafizur; Bouguila, Nizar.

2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020. Vol. 2020-October Institute of Electrical and Electronics Engineers Inc., 2020. p. 3588-3593 9283063.

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

Rahman, MH & Bouguila, N 2020, Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification. in 2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020. vol. 2020-October, 9283063, Institute of Electrical and Electronics Engineers Inc., pp. 3588-3593, 2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020, Toronto, Canada, 10/11/20. https://doi.org/10.1109/SMC42975.2020.9283063

@inproceedings{8b9b40b168aa4ec0add1300467954748,

title = "Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification",

abstract = "In this paper, we propose a distribution based feature mapping technique to improve the baseline accuracy of SVM kernels in different computer vision tasks. The proposed technique is based on learning parameters from the data and use that parameters to make inference from new data. The learned parameters can be thought of as prior knowledge about the data representation. Utilizing such prior knowledge about the data distribution increases the discriminative power of the classifier. Our proposed feature mapping technique is based on inverted Dirichlet, generalized inverted Dirichlet and inverted Beta Liouville distributions. These distributions are efficient in modelling semi-bounded data which are prevalent in computer vision problems. Our experimental results demonstrate the effectiveness of the proposed method in texture recognition, natural scene recognition and human action recognition in videos.",

keywords = "Feature Mapping, Generalized Inverted Dirichlet Distribution, Inverted Beta-Liouville Distribution, Inverted Dirichlet Distribution, SVM",

author = "Rahman, {Md Hafizur} and Nizar Bouguila",

note = "Funding Information: The completion of this research was made possible thanks to the Natural Sciences and Engineering Research Council of Canada (NSERC).; 2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020 ; Conference date: 11-10-2020 Through 14-10-2020",

year = "2020",

month = oct,

day = "11",

doi = "10.1109/SMC42975.2020.9283063",

language = "English (US)",

volume = "2020-October",

pages = "3588--3593",

booktitle = "2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Inverted Dirichlet and Related Distributions Based Feature Mapping for Data Classification

AU - Rahman, Md Hafizur

AU - Bouguila, Nizar

N1 - Funding Information: The completion of this research was made possible thanks to the Natural Sciences and Engineering Research Council of Canada (NSERC).

PY - 2020/10/11

Y1 - 2020/10/11

N2 - In this paper, we propose a distribution based feature mapping technique to improve the baseline accuracy of SVM kernels in different computer vision tasks. The proposed technique is based on learning parameters from the data and use that parameters to make inference from new data. The learned parameters can be thought of as prior knowledge about the data representation. Utilizing such prior knowledge about the data distribution increases the discriminative power of the classifier. Our proposed feature mapping technique is based on inverted Dirichlet, generalized inverted Dirichlet and inverted Beta Liouville distributions. These distributions are efficient in modelling semi-bounded data which are prevalent in computer vision problems. Our experimental results demonstrate the effectiveness of the proposed method in texture recognition, natural scene recognition and human action recognition in videos.

AB - In this paper, we propose a distribution based feature mapping technique to improve the baseline accuracy of SVM kernels in different computer vision tasks. The proposed technique is based on learning parameters from the data and use that parameters to make inference from new data. The learned parameters can be thought of as prior knowledge about the data representation. Utilizing such prior knowledge about the data distribution increases the discriminative power of the classifier. Our proposed feature mapping technique is based on inverted Dirichlet, generalized inverted Dirichlet and inverted Beta Liouville distributions. These distributions are efficient in modelling semi-bounded data which are prevalent in computer vision problems. Our experimental results demonstrate the effectiveness of the proposed method in texture recognition, natural scene recognition and human action recognition in videos.

KW - Feature Mapping

KW - Generalized Inverted Dirichlet Distribution

KW - Inverted Beta-Liouville Distribution

KW - Inverted Dirichlet Distribution

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=85098887619&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85098887619&partnerID=8YFLogxK

U2 - 10.1109/SMC42975.2020.9283063

DO - 10.1109/SMC42975.2020.9283063

M3 - Conference contribution

AN - SCOPUS:85098887619

VL - 2020-October

SP - 3588

EP - 3593

BT - 2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2020

Y2 - 11 October 2020 through 14 October 2020

ER -