Login | Register

Statistical modeling for simultaneous data clustering, features selection, and outliers rejection

Title:

Statistical modeling for simultaneous data clustering, features selection, and outliers rejection

Almakadmeh, Khaled (2010) Statistical modeling for simultaneous data clustering, features selection, and outliers rejection. Masters thesis, Concordia University.

[img]
Preview
Text (application/pdf)
MR71008.pdf - Accepted Version
2MB

Abstract

Model-based approaches and in particular finite mixture models are widely used for data clustering, which is a crucial step in several applications of practical importance. Indeed, many pattern recognition, computer vision, and image processing applications can be approached as feature space clustering problems. However, the use of these approaches for complex high-dimensional data presents several challenges such as the presence of many irrelevant features, which may affect the speed, and compromise the accuracy of the used learning algorithm. Another problem is the presence of outliers which potentially influence the resulting model parameters. Generally; clustering, features selection, and outliers detection problems have been approached separately. In this thesis, we propose a unified statistical framework to address the three problems simultaneously. The proposed statistical model partitions a given data set without a priori information about the number of clusters, the saliency of the features, or the number of outliers. We illustrate the performance of our approach using different applications involving synthetic data, real data, and objects shape clustering.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Concordia Institute for Information Systems Engineering
Item Type:Thesis (Masters)
Authors:Almakadmeh, Khaled
Pagination:xi, 59 leaves : ill. ; 29 cm.
Institution:Concordia University
Degree Name:M.A. Sc.
Program:Institute for Information Systems Engineering
Date:2010
Thesis Supervisor(s):Bouguila, Nizar
ID Code:979461
Deposited By: Concordia University Library
Deposited On:09 Dec 2014 17:59
Last Modified:18 Jan 2018 17:49
Related URLs:
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Back to top Back to top