Deterministic Annealing Multi-sphere Support Vector Data Description

Trung Le, Dat Tran, Wanli Ma, Dharmendra Sharma

Research output: A Conference proceeding or a Chapter in BookConference contributionpeer-review

1 Citation (Scopus)
3 Downloads (Pure)

Abstract

Current well-known data description method such as Support Vector Data Description is conducted with assumption that data samples of a class in feature space are drawn from a single distribution. Based on this assumption, a single hypersphere is constructed to provide a good data description for the data. However, real-world data samples may be drawn from some distinctive distributions and hence it does not guarantee that a single hypersphere can offer the best data description. In this paper, we introduce a Deterministic Annealing Multi-sphere Support Vector Data Description (DAMS-SVDD) approach to address this issue. We propose to use a set of hyperspheres to provide a better data description for a given data set. Calculations for determining optimal hyperspheres and experimental results for applying this proposed approach to classification problems are presented.
Original languageEnglish
Title of host publicationInternational Conference on Neural Information Processing
Subtitle of host publicationLecture Notes in Computer Science
EditorsTingwen Huang, Zhigang Zeng, Chuandong Li, Chi Sing Leung
Place of PublicationBerlin Heidellberg, Germany
PublisherSpringer
Pages183-190
Number of pages8
Volume7665
ISBN (Print)9783642344879
DOIs
Publication statusPublished - 2012
Event19th International Conference on Neural Information Processing 2012 - Doha, Doha, Qatar
Duration: 12 Nov 201215 Nov 2012

Conference

Conference19th International Conference on Neural Information Processing 2012
Country/TerritoryQatar
CityDoha
Period12/11/1215/11/12

Fingerprint

Dive into the research topics of 'Deterministic Annealing Multi-sphere Support Vector Data Description'. Together they form a unique fingerprint.

Cite this