A novel lie detection method based on extreme learning machine using P300

• Published in: ICISCE 2012 : IET International Conference on Information Science and Control Engineering 2012 : 7-9 December 2012 p. 3.74
• Main Authors: Xiong, Yijun, Yang, Yong, Gao, Junfeng
• Format: Conference Proceeding
• Language: English
• Published: Stevenage, UK IET 2012; The Institution of Engineering & Technology
• Subjects: Bioelectric signals; Biology and medical computing; Digital signal processing; Electrical activity in neurophysiological processes; Knowledge engineering techniques; Neural computing techniques; support vector machines; bioelectric potentials; single layer feedforward network; support vector machine classifier; SLFN; ELM; lie detection method; P300 potential; medical signal processing; backpropagation network; extreme learning machine; backpropagation; feedforward neural nets
• ISBN: 9781849196413; 1849196419
• DOI: 10.1049/cp.2012.2471

Machine learning-based lie detection has drawn much attention recently. In this paper, we used extreme learning machine (ELM), a recently-proposed machine learning method based on a single layer feedforward network (SLFN), to classify P300 potentials from guilty subject and non-P300 potentials from innocent subject. Back-propagation network and support vector machine classifiers were also used to compare with the proposed method. The number of hidden nodes in ELM was tuned using training with the 10-fold cross validation. The experimental results show that the proposed method reaches the highest classification accuracy with extremely less training and testing time, compared with the other classification models.