Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Jun 3;8(6):e64704.
doi: 10.1371/journal.pone.0064704. Print 2014.

A novel approach for lie detection based on F-score and extreme learning machine

Junfeng Gao  1 , Zhao Wang, Yong Yang, Wenjia Zhang, Chunyi Tao, Jinan Guan, Nini Rao
Affiliations

A novel approach for lie detection based on F-score and extreme learning machine

Junfeng Gao et al. PLoS One. .

Abstract

A new machine learning method referred to as F-score_ELM was proposed to classify the lying and truth-telling using the electroencephalogram (EEG) signals from 28 guilty and innocent subjects. Thirty-one features were extracted from the probe responses from these subjects. Then, a recently-developed classifier called extreme learning machine (ELM) was combined with F-score, a simple but effective feature selection method, to jointly optimize the number of the hidden nodes of ELM and the feature subset by a grid-searching training procedure. The method was compared to two classification models combining principal component analysis with back-propagation network and support vector machine classifiers. We thoroughly assessed the performance of these classification models including the training and testing time, sensitivity and specificity from the training and testing sets, as well as network size. The experimental results showed that the number of the hidden nodes can be effectively optimized by the proposed method. Also, F-score_ELM obtained the best classification accuracy and required the shortest training and testing time.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. The preprocessing results of a guilty subject and an innocent subject.
1A: Three averaged waves over the three kinds of stimuli respectively at Pz electrode from the guilty subject. 1B: The averaged waves over the three kinds of stimuli respectively at Pz electrode from the innocent subject. 1C: The brain topographies at the latency of 348 ms of the averaged P responses (the solid line in Figure 1A). 1D: The brain topographies at the peak point of 316 ms of the averaged P responses (the solid line in Figure 1B).
Figure 2
Figure 2. SLFN with K hidden, n input and m output nodes.
Figure 3
Figure 3. The block diagram of the proposed method F-score_ELM.
Figure 4
Figure 4. Training accuracy and NHN/NSV as a function of NFS achieved by the three classification models.
Each point in each curve corresponds to the highest classification performance of the indicated model with the optimal NHN/NSV. 4A: Highest sensitivity with the optimal NHN or NSV vs NFS. 4B: Highest specificity with the optimal NHN or NSV vs NFS. 4C: NHN vs NFS for which BA_train achieves its highest value. 4D: NSV vs NFS for which BA_train achieves its highest value.
Figure 5
Figure 5. Training accuracy (constant NFS = 11) as a function of NHN achieve by the F-score_ELM.
5A: Highest sensitivity formula image vs log (NHN). 5B: Highest specificity formula image vs log (NHN).
See this image and copyright information in PMC

References

    1. Langleben DD, Loughead JW, Bilker WB, Ruparel K, Childress AR, et al. (2005) Telling truth from lie in individual subjects with fast event-related fMRI. Hum Brain Mapp 26(4): 262–272. - PMC - PubMed
    1. Phan KL, Magalhaes A, Ziemlewicz TJ, Fitzgerald DA, Green C, et al. (2005) Neural correlates of telling lies: a functional magnetic resonance imaging study at 4Tesla. Acad Radiol 12(2): 164–172. - PubMed
    1. Rosenfeld JP (2002) Event-related potentials in the detection of deception. Handbook of Polygraph Testing. Academic Press, New York. 265–286.
    1. Davatzikos C, Ruparel K, Fan Y, Shen DG, Acharyya M, et al. (2005) Classifying spatial patterns of brain activity with machine learning methods: Application to lie detection. NeuroImage 28(3): 663–668. - PubMed
    1. Polich J, Herbst KL (2000) P300 as a clinical assay: rational, evaluation, and findings. Int J Psychophysi 38(1): 3–19. - PubMed

Publication types

LinkOut - more resources