publication . Conference object . 2016

AKSDA-MSVM

Hanjalic, Alan; Snoek, Cees; Worring, Marcel; Bulterman, Dick; Huet, Benoit; Kelliher, Aisling; Kompatsiaris, Yiannis; Li, Jin; Arestis-Chartampilas, Stavros; Gkalelis, Nikolaos; ...
Open Access English
  • Published: 17 Oct 2016
  • Publisher: ACM
Abstract
In this paper, a combined nonlinear dimensionality reduction and multiclass classification framework is proposed. Specifically, a novel discriminant analysis (DA) technique, called accelerated kernel subclass discriminant analysis (AKSDA), derives a discriminant subspace, and a linear multiclass support vector machine (MSVM) computes a set of separating hyperplanes in the derived subspace. Moreover, within this framework an approach for accelerating the computation of multiple Gram matrices and an associated late fusion scheme are presented. Experimental evaluation in five multimedia datasets, on tasks such as video event detection and news document classificati...
Persistent Identifiers
Subjects
ACM Computing Classification System: ComputingMethodologies_PATTERNRECOGNITION
free text keywords: Discriminant analysis; GPU; multiclass classification; SVM, Multimedia, computer.software_genre, computer, Artificial intelligence, business.industry, business, Multiple discriminant analysis, Linear discriminant analysis, Machine learning, Nonlinear dimensionality reduction, Kernel (linear algebra), Support vector machine, Pattern recognition, Optimal discriminant analysis, Multiclass classification, Subspace topology, Computer science
Funded by
EC| InVID
Project
InVID
In Video Veritas – Verification of Social Media Video Content for the News Industry
  • Funder: European Commission (EC)
  • Project Code: 687786
  • Funding stream: H2020 | IA
Validated by funder
,
EC| FORGETIT
Project
FORGETIT
Concise Preservation by combining Managed Forgetting and Contextualized Remembering
  • Funder: European Commission (EC)
  • Project Code: 600826
  • Funding stream: FP7 | SP1 | ICT
Any information missing or wrong?Report an Issue