# Multiclass classification using Least Squares Support Vector Machine
> Jafar N.

URL kanonis: https://discover.unhas.ac.id/publications/multiclass-classification-using-least-squares-support-vector-machine
Jurnal / Konferensi: Proceedings Cyberneticscom 2016 International Conference on Computational Intelligence and Cybernetics
Tahun terbit: 2017
DOI: https://doi.org/10.1109/CyberneticsCom.2016.7892558
Citations: 6

## Authors
- Jafar N.

## Abstract
In this paper, multiclass classification problem; One Against All and One Against One, with Least Squares Support Vector Machine (LS-SVM) will be used. There are three type of kernels were used in this paper; Radial Basis Function (RBF), polynomial and linear. One Against All method and One Against One method will be compared to see the accuracy of each kernel, and the amount of misclassification using the confusion matrix. This is illustrated by using iris plant species dataset and the preferred method of contraception dataset. The results showed that the method of One Against One is better than the One Against All based on the accuracy for kernels RBF, polynomial, and linear.

## Keywords
- Support vector machine
- Least squares support vector machine
- Polynomial kernel
- Pattern recognition (psychology)
- Kernel (algebra)
- Artificial intelligence
- Confusion matrix
- Radial basis function
- Computer science
- Radial basis function kernel
- Least-squares function approximation
- Polynomial
- Relevance vector machine
- Mathematics
- Multiclass classification
- Kernel method
- Artificial neural network
- Statistics
- Mathematical analysis
- Estimator
- Combinatorics

---
Sumber: Discover Unhas — RIMS Universitas Hasanuddin.
Saat mengutip, gunakan DOI bila tersedia atau URL kanonis di atas.