Performance Evaluation Of Support Vector Machines (Svms)And Convolutional Neural Network (Cnn) On Binary Classification Problem
Journal Title: International Journal of engineering Research and Applications - Year 2018, Vol 8, Issue 9
Abstract
Support vector machines (SVMs) have been around for decades, they have been used for a number of classification tasks. They actually have a very strong theory behind them, which make it relatively easy to choose the best hyper-parameters. The kernel trick makes it easier for SVMs to implicitly classify in higher dimensional space, making it possible to work with nonlinearly separable datasets. On the other hand, Convolutional Neural Networks (CNNs) have gained important attention in recent years for their high performance in image classification problems with high number of categories. The automatic feature extraction of convolutional layer and the dimensionality reduction of the pooling layer make CNN gain high predictive power on testing data. In this work both models are briefly discussed and implemented on a binary classification problem from the EMIST character dataset. The CNN outperformed SVM achieving a misclassification error rate on test data of 1.7 % against 2.32 % for SVM.
Authors and Affiliations
Abdelaziz Botalb
Keywords
Related Articles
- EP ID EP395176
- DOI 10.9790/9622-0809044448.
- Views 129
- Downloads 0
Integration of Finite Element Method with Runge – Kuta Solution Algorithm
Runge – Kuta (RK) method is reasonably simple and robust for numerical solution of differential equations but it requires an intelligent adaptive step-size routine; to achieve this, there is need to develop a good logica...
Fuzzy Matrix for Ebola Virus Disease in Human Body
The aim of this paper is to design a Fuzzy soft matrix for Ebola virus disease diagnosis, by using the notion of fuzzy soft matrices. We apply fuzzy soft set technology through the well known Sanchez’s approach for HUMAN...
Optimizing TCP Congestion Control techniques for modern Wireless Network
Transmission Control Protocol (TCP) intended to convey consistent and solid end-to-end information exchange crosswise over untrustworthy systems works perfectly well in wired condition. Indeed, TCP bears the 90% of Inter...
Link Conflict in Multistage Interconnection Networks
Multistage interconnection networks (MINs) are high performance indispensable networks with low cost for broadband switching technology and multiprocessor systems. Crosstalk is a critical issue in MINs. The undesired cou...
Prediction of Two StageAir Compressor Characteristic
The paper reports investigation of indicating power input, the poly topic compression index and overall efficiency of two stage air compressor under condition of atmospheric pressure and temperature in Shuwaikh Kuwait. T...