Gender classification by principal component analysis and support vector machine
Sunita Kumari, Pankaj Kumar, Banshidhar Majhi · 2011
Performance of any system is identified by its accuracy and speed. Accuracy depends on underlying algorithm while speed depends on size of the database. A tradeoff between these two contradictory aspects has to be achieved. This paper addresses the problem of speed using gender classification. Principal Component AnalysisPCA is used to represent each image as a feature vector in a low dimensional subspace and then a non-linear Support Vector Machine(SVM) is used for gender classification. Experimental results show an accuracy of 92% and is compared with other existing research.