Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION

Abstract: 

Pattern recognition on big data can be challenging for kernel machines as the complexity grows with the squared number of training samples. In this work, we overcome this hurdle via the outlying data sample removal pre-processing step. This approach removes less informative data samples and trains the kernel machines only with the remaining data, and hence, directly reduces the complexity by reducing the number of training samples. To enhance the classification performance, the outlier removal process is done such that the discriminant information of the data is mostly intact. This is achieved via the novel Outlier-Removal Discriminant Information (ORDI) metric, which measures the contribution of each sample toward the discriminant information of the dataset. Hence, the ORDI metric can be used together with the simple filter method to effectively remove insignificant outliers to both reduce the computational cost and enhance the classification performance. We experimentally show on two real-world datasets at the sample removal ratio of 0.2 that, with outlier removal via ORDI, we can simultaneously (1) improve the accuracy of the classifier by 1%, and (2) provide significant saving on the total running time by 1.5x and 2x on the two datasets. Hence, ORDI can provide a win-win situation in this performance-complexity tradeoff of the kernel machines for big data analysis.

ordi2.pdf

PDF icon ordi2.pdf (6 downloads)
up
0 users have voted:

Paper Details

Authors:
Thee Chanyaswad, Mert Al, Sun-Yuan Kung
Submitted On:
14 April 2018 - 9:10pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Mert Al
Paper Code:
MLSP-L4.2
Document Year:
2018
Cite

Document Files

ordi2.pdf

(6 downloads)

Subscribe

[1] Thee Chanyaswad, Mert Al, Sun-Yuan Kung, "OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2754. Accessed: Apr. 26, 2018.
@article{2754-18,
url = {http://sigport.org/2754},
author = {Thee Chanyaswad; Mert Al; Sun-Yuan Kung },
publisher = {IEEE SigPort},
title = {OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION},
year = {2018} }
TY - EJOUR
T1 - OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION
AU - Thee Chanyaswad; Mert Al; Sun-Yuan Kung
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2754
ER -
Thee Chanyaswad, Mert Al, Sun-Yuan Kung. (2018). OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION. IEEE SigPort. http://sigport.org/2754
Thee Chanyaswad, Mert Al, Sun-Yuan Kung, 2018. OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION. Available at: http://sigport.org/2754.
Thee Chanyaswad, Mert Al, Sun-Yuan Kung. (2018). "OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION." Web.
1. Thee Chanyaswad, Mert Al, Sun-Yuan Kung. OUTLIER REMOVAL FOR ENHANCING KERNEL-BASED CLASSIFIER VIA THE DISCRIMINANT INFORMATION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2754