ACM Transactions on Knowledge Discovery from Data (TKDD)
Evaluating multi-class multiple-instance learning for image categorization
ACCV'07 Proceedings of the 8th Asian conference on Computer vision - Volume Part II
Hi-index | 0.00 |
Binary Support Vector Machines (SVM) have proven effec- tive in classiffication. However, problems remain with respect to feature selection in multi-class classification. This article proposes a novel multi-class SVM, which performs classifica- tion and feature selection simultaneously via L_1-norm penal- ized sparse representations. The proposed methodology, to- gether with our developed regularization solution path, per- mits feature selection within the framework of classiffication. The operational characteristics of the proposed methodol- ogy is examined via both simulated and benchmark exam- ples, and is compared to some competitors in terms of the accuracy of prediction and feature selection. The numeri- cal results suggest that the proposed methodology is highly competitive.