A New Fuzzy Comprehensive Evaluation Model Based on the Support Vector Machine
Author:
- 1 March 2010
- journal article
- Published by Tsinghua University Press in Fuzzy Information and Engineering
- Vol. 2 (1) , 75-86
Abstract
In this paper, focused on a proper weight distribution vector of a fuzzy comprehensive evaluation method, a new fuzzy comprehensive evaluation model based on the supper vector machine (SVM) is proposed for overcoming the subjective limitation in traditional fuzzy comprehensive evaluation model. Furthermore, a multilevel fuzzy comprehensive evaluation model based on SVM of network learning has also been designed, and the improved algorithm is used to make an instant computation. This method gives good performance on determination of the weight distribution vector and improves the evaluation accuracy and generalization with an example.Keywords
This publication has 8 references indexed in Scilit:
- An improved multilevel fuzzy comprehensive evaluation algorithm for security performanceThe Journal of China Universities of Posts and Telecommunications, 2006
- Feature Selection using Fuzzy Support Vector MachinesFuzzy Optimization and Decision Making, 2006
- Learning with Kernels2001
- An introduction to kernel-based learning algorithmsIEEE Transactions on Neural Networks, 2001
- Tolerance synthesis using second-order fuzzy comprehensive evaluation and genetic algorithmInternational Journal of Production Research, 2000
- Support-vector networksMachine Learning, 1995
- The Nature of Statistical Learning Theory1995
- Single-layer learning revisited: a stepwise procedure for building and training a neural network1990