Parallel Randomized Support Vector Machine

A parallel support vector machine based on randomized sampling technique is proposed in this paper. We modeled a new LP-type problem so that it works for general linear-nonseparable SVM training problems unlike the previous work [2]. A unique priority based sampling mechanism is used so that we can...

Full description

Saved in:
Bibliographic Details
Published inAdvances in Knowledge Discovery and Data Mining pp. 205 - 214
Main Authors Lu, Yumao, Roychowdhury, Vwani
Format Book Chapter Conference Proceeding
LanguageEnglish
Published Berlin, Heidelberg Springer Berlin Heidelberg 2006
Springer
SeriesLecture Notes in Computer Science
Subjects
Applied sciences
Combinatorial Dimension
Computer science; control theory; systems
Data processing. List processing. Character string processing
Exact sciences and technology
Geographic Information System Database
Information systems. Data bases
Memory organisation. Data processing
Regularization Factor
Software
Support Vector Machine
Training Vector
Data analysis
Statistical analysis
Scalability
Parallel machines
Linear programming
Data mining
Modeling
Randomization
Numerical data
Database
Vector support machine
Convergence rate
Knowledge discovery
Sampling
Online AccessGet full text

Cover

More Information
Summary:A parallel support vector machine based on randomized sampling technique is proposed in this paper. We modeled a new LP-type problem so that it works for general linear-nonseparable SVM training problems unlike the previous work [2]. A unique priority based sampling mechanism is used so that we can prove an average convergence rate that is so far the fastest bounded convergence rate to the best of our knowledge. The numerical results on synthesized data and a real geometric database show that our algorithm has good scalability.
ISBN:9783540332060
3540332065
ISSN:0302-9743
1611-3349
DOI:10.1007/11731139_25