Lars Lundberg
ABSTRACT
We present a new method, called hyperplane folding, that increases the margin in Support Vector Machines (SVMs). Based on the location of the support vectors, the method splits the dataset into two parts, rotates one part of the dataset and then merges the two parts again. This procedure increases the margin as long as the margin is smaller than half of the shortest distance between any pair of data points from the two different classes. We provide an algorithm for the general case with n-dimensional data points. A small experiment with three folding iterations on 3-dimensional data points with non-linear relations shows that the margin does indeed increase and that the accuracy improves with a larger margin. The method can use any standard SVM implementation plus some basic manipulation of the data points, i.e., splitting, rotating and merging. Hyperplane folding also increases the interpretability of the data.
- Ali, S., Smith-Miles, K.A.: A meta-learning approach to automatic kernel selection for support vector machines. Neurocomputing 70 (2006) 173--186.Google Scholar
Cross Ref
- Boeva, V., Lundberg, L., Angelova, M., Kohstall, J.: Cluster Validation Measures for Label Noise Filtering, The 9th IEEE International Conference on Intelligent Systems IS'18, Madeira Island, Portugal, September 25--27, 2018.Google ScholarCross Ref
- Cortes, C., Vapnik, V.: Support-vector networks. Machine learning 20(3) (1995) 273--297. Google ScholarDigital Library
- Gill, N., Hall, P.: An Introduction to Machine Learning Interpretability. O'Reilly Media Inc., April 2018, ISBN: 9781492033141.Google Scholar
- Golino, H.F., de Brito Amaral, L.S., Duarte, S.F.P., Gomes, C.M.A., de Jesus Soares, T., dos Reis, L.A., Santos, J.: Predicting Increased Blood Pressure Using Machine Learning. Journal of Obesity, Volume 2014, Article ID 637635.Google Scholar
- Huang, X., Mehrkanoon, S., Suykens, J.A.: Support vector machines with piecewise linear feature mapping. Neurocomputing 117 (2013) 118--127.Google ScholarCross Ref
- Kostin, A.: A simple and fast multi-class piecewise linear pattern classifier. Pattern Recognition 39 (2006) 1949--1962. Google ScholarDigital Library
- Li, Y., Leng, Q., Fu, Y., Li, H.: Growing construction of conlitron and multiconlitron. Knowledge-Based Systems 65 (2014) 12--20. Google ScholarDigital Library
- Nedaie, A., Najafi, A.A.: Support vector machine with Dirichlet feature mapping. Neural Networks 98 (2018) 87--101.Google ScholarCross Ref
- Sklansky, J., Michelotti, L.: Locally trained piecewise linear classifiers, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 2, No. 2 (1980) 101--111. Google ScholarDigital Library
- Tenmoto, H., Kudo, M., Shimbo, M.: Piecewise linear classifiers with an appropriate number of hyperplanes. Pattern Recognition, Vol 31, No 11 (1998) 1627--1634.Google ScholarCross Ref
- Vapnik, V.: The nature of statistical learning theory. Springer (1995). Google ScholarDigital Library
- Wang, S., Sun, X.: Generalization of hinging hyperplanes. IEEE Transactions on Information Theory 51(12) (2005) 4425--4431. Google ScholarDigital Library
- Yujian, L., Bo, L., Xinwu, Y., Yaozong, F., Houjun, L.: Multiconlitron: A general piecewise linear classifier. IEEE Transactions on Neural Networks 22(2) (2011) 276--289. Google ScholarDigital Library
- Yujian, L., Qiangkui, L.: Alternating multiconlitron: A novel framework for piecewise linear classification. Pattern Recognition 48 (2015) 968--975. Google ScholarDigital Library
Index Terms
- Handling Non-linear Relations in Support Vector Machines through Hyperplane Folding
Recommendations
An overview on twin support vector machines
Twin support vector machines (TWSVM) is based on the idea of proximal SVM based on generalized eigenvalues (GEPSVM), which determines two nonparallel planes by solving two related SVM-type problems, so that its computing cost in the training phase is 1/...
Incremental training of support vector machines using hyperspheres
In the conventional incremental training of support vector machines, candidates for support vectors tend to be deleted if the separating hyperplane rotates as the training data are added. To solve this problem, in this paper, we propose an incremental ...
Twin Support Vector Machines for Pattern Classification
We propose Twin SVM, a binary SVM classifier that determines two nonparallel planes by solving two related SVM-type problems, each of which is smaller than in a conventional SVM. The Twin SVM formulation is in the spirit of proximal SVMs via generalized ...
Comments