Fast training of support vector machines using sequential minimal optimization
Advances in kernel methods
Learning and Soft Computing: Support Vector Machines, Neural Networks, and Fuzzy Logic Models
Learning and Soft Computing: Support Vector Machines, Neural Networks, and Fuzzy Logic Models
In Defense of One-Vs-All Classification
The Journal of Machine Learning Research
Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised, and Unsupervised Learning (Studies in Computational Intelligence)
| Hi-index | 0.00 |
The paper presents the results of using Support Vector Machines (SVMs) and Radial Basis Function Neural Networks (RBF NNs) for diagnosing erythemato-squamous diseases which represent difficult dermatological problems. The data sets contains 358 data pairs of 34 dimensional input records of patients with six known diagnosis (outputs). Thus, data set is sparse and it is fairly unbalanced too. The paper also discusses the strategies for training SVMs. Both networks design six different one-against-other classifier models which show extremely good performance on previously unseen test data. The training and the test sets are obtained by random splitting the dataset into two groups ensuring that each group contains at least one patient for each disease. 100 random split trials (equivalent to performing 10-fold-crossvalidation 10 times independently) were carried out for estimating the tests error rates.