Matches in SemOpenAlex for { <https://semopenalex.org/work/W2134767360> ?p ?o ?g. }
- W2134767360 endingPage "8005" @default.
- W2134767360 startingPage "7964" @default.
- W2134767360 abstract "Traditional extensions of the binary support vector machine (SVM) to multiclass problems are either heuristics or require solving a large dual optimization problem. Here, a generalized multiclass SVM is proposed called GenSVM. In this method classification boundaries for a K-class problem are constructed in a (K - 1)-dimensional space using a simplex encoding. Additionally, several different weightings of the misclassification errors are incorporated in the loss function, such that it generalizes three existing multiclass SVMs through a single optimization problem. An iterative majorization algorithm is derived that solves the optimization problem without the need of a dual formulation. This algorithm has the advantage that it can use warm starts during cross validation and during a grid search, which significantly speeds up the training phase. Rigorous numerical experiments compare linear GenSVM with seven existing multiclass SVMs on both small and large data sets. These comparisons show that the proposed method is competitive with existing methods in both predictive accuracy and training time, and that it significantly outperforms several existing methods on these criteria." @default.
- W2134767360 created "2016-06-24" @default.
- W2134767360 creator A5062156497 @default.
- W2134767360 creator A5066593623 @default.
- W2134767360 date "2016-01-01" @default.
- W2134767360 modified "2023-09-30" @default.
- W2134767360 title "GenSVM: a generalized multiclass support vector machine" @default.
- W2134767360 cites W1480376833 @default.
- W2134767360 cites W1494922272 @default.
- W2134767360 cites W1512098439 @default.
- W2134767360 cites W1523794535 @default.
- W2134767360 cites W1526146785 @default.
- W2134767360 cites W1536816303 @default.
- W2134767360 cites W1556598185 @default.
- W2134767360 cites W1565746575 @default.
- W2134767360 cites W1573195610 @default.
- W2134767360 cites W1602492977 @default.
- W2134767360 cites W1676820704 @default.
- W2134767360 cites W187370165 @default.
- W2134767360 cites W1974758710 @default.
- W2134767360 cites W2000950277 @default.
- W2134767360 cites W2010486434 @default.
- W2134767360 cites W2019162816 @default.
- W2134767360 cites W2035720976 @default.
- W2134767360 cites W2036001960 @default.
- W2134767360 cites W2046033161 @default.
- W2134767360 cites W2059392350 @default.
- W2134767360 cites W2059919683 @default.
- W2134767360 cites W2073848658 @default.
- W2134767360 cites W2077139171 @default.
- W2134767360 cites W2090963365 @default.
- W2134767360 cites W2101276256 @default.
- W2134767360 cites W2112081648 @default.
- W2134767360 cites W2112545207 @default.
- W2134767360 cites W2116444583 @default.
- W2134767360 cites W2116915136 @default.
- W2134767360 cites W2118585731 @default.
- W2134767360 cites W2119821739 @default.
- W2134767360 cites W2121044470 @default.
- W2134767360 cites W2125993116 @default.
- W2134767360 cites W2133864802 @default.
- W2134767360 cites W2134426089 @default.
- W2134767360 cites W2135653967 @default.
- W2134767360 cites W2146851241 @default.
- W2134767360 cites W2147898188 @default.
- W2134767360 cites W2148603752 @default.
- W2134767360 cites W2149414429 @default.
- W2134767360 cites W2149454052 @default.
- W2134767360 cites W2151922881 @default.
- W2134767360 cites W2153104898 @default.
- W2134767360 cites W2153635508 @default.
- W2134767360 cites W2154776925 @default.
- W2134767360 cites W2157239837 @default.
- W2134767360 cites W2157791002 @default.
- W2134767360 cites W2165966284 @default.
- W2134767360 cites W2166879083 @default.
- W2134767360 cites W2167197286 @default.
- W2134767360 cites W2172000360 @default.
- W2134767360 cites W2172219851 @default.
- W2134767360 cites W2182722412 @default.
- W2134767360 cites W2187010386 @default.
- W2134767360 cites W2228493906 @default.
- W2134767360 cites W2282078507 @default.
- W2134767360 cites W2333672468 @default.
- W2134767360 cites W2546945975 @default.
- W2134767360 cites W3120740533 @default.
- W2134767360 cites W410724028 @default.
- W2134767360 cites W56897548 @default.
- W2134767360 hasPublicationYear "2016" @default.
- W2134767360 type Work @default.
- W2134767360 sameAs 2134767360 @default.
- W2134767360 citedByCount "7" @default.
- W2134767360 countsByYear W21347673602017 @default.
- W2134767360 countsByYear W21347673602018 @default.
- W2134767360 countsByYear W21347673602019 @default.
- W2134767360 countsByYear W21347673602020 @default.
- W2134767360 countsByYear W21347673602021 @default.
- W2134767360 crossrefType "journal-article" @default.
- W2134767360 hasAuthorship W2134767360A5062156497 @default.
- W2134767360 hasAuthorship W2134767360A5066593623 @default.
- W2134767360 hasConcept C11413529 @default.
- W2134767360 hasConcept C119857082 @default.
- W2134767360 hasConcept C12267149 @default.
- W2134767360 hasConcept C123860398 @default.
- W2134767360 hasConcept C125168437 @default.
- W2134767360 hasConcept C126255220 @default.
- W2134767360 hasConcept C127705205 @default.
- W2134767360 hasConcept C154945302 @default.
- W2134767360 hasConcept C33923547 @default.
- W2134767360 hasConcept C41008148 @default.
- W2134767360 hasConcept C66905080 @default.
- W2134767360 hasConceptScore W2134767360C11413529 @default.
- W2134767360 hasConceptScore W2134767360C119857082 @default.
- W2134767360 hasConceptScore W2134767360C12267149 @default.
- W2134767360 hasConceptScore W2134767360C123860398 @default.
- W2134767360 hasConceptScore W2134767360C125168437 @default.
- W2134767360 hasConceptScore W2134767360C126255220 @default.
- W2134767360 hasConceptScore W2134767360C127705205 @default.