Please use this identifier to cite or link to this item:
|Title:||Nomogram visualization for ranking support vector machine||Authors:||Thuy, N.T.T.
probabilistic ranking SVM
|Issue Date:||2011||Citation:||Thuy, N.T.T.,Vinh, N.T.N.,Vien, N.A. (2011). Nomogram visualization for ranking support vector machine. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 6676 LNCS (PART 2) : 94-102. ScholarBank@NUS Repository. https://doi.org/10.1007/978-3-642-21090-7_11||Abstract:||In this paper, we propose a visualization model for a trained ranking support vector machine. In addition, we also introduce a feature selection method for the ranking support vector machine, and show visually each feature's effect. Nomogram is a well-known visualization model that graphically describes completely the model on a single graph. The complexity of the visualization does not depend on the number of the features but on the properties of the kernel. In order to represent the effect of each feature on the log odds ratio on the nomograms, we use probabilistic ranking support vector machines which map the support vector machine outputs into a probabilistic sigmoid function whose parameters are trained by using cross-validation. The experiments show the effectiveness of our proposal which helps the analysts study the effects of predictive features. © 2011 Springer-Verlag.||Source Title:||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)||URI:||http://scholarbank.nus.edu.sg/handle/10635/41589||ISBN:||9783642210891||ISSN:||03029743||DOI:||10.1007/978-3-642-21090-7_11|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.