A Comparison of Classification Methods for Predicting Deception in Computer-Mediated Communication.

Author List: Zhou, Lina; Burgoon, Judee K.; TWITCHELL, DOUGLAS P.; QIN, TIANTIAN; Nunamaker, Jr., Jay F.;
Journal of Management Information Systems, 2004, Volume 20, Issue 4, Page 139-165.
The increased chance of deception in computer-mediated communication and the potential risk of taking action based on deceptive information calls for automatic detection of deception. To achieve the ultimate goal of automatic prediction of deception, we selected four common classification methods and empirically compared their performance in predicting deception. The deception and truth data were collected during two experimental studies. The results suggest that all of the four methods were promising for predicting deception with cues to deception. Among them, neural networks exhibited consistent performance and were robust across test settings. The comparisons also highlighted the importance of selecting important input variables and removing noise in an attempt to enhance the performance of classification methods. The selected cues offer both methodological and theoretical contributions to the body of deception and information systems research.
Keywords: classification methods; deception; deception detection; linguistic cues
Algorithm:

List of Topics

#7 0.378 detection deception assessment credibility automated fraud fake cues detecting results screening study detect design indicators science important theory performance improved
#215 0.303 data classification statistical regression mining models neural methods using analysis techniques performance predictive networks accuracy method variables prediction problem measure
#108 0.066 model research data results study using theoretical influence findings theory support implications test collected tested based empirical empirically context paper
#203 0.051 communication media computer-mediated e-mail richness electronic cmc mail medium message performance convergence used communications messages face-to-face findings participants results work
#204 0.051 goals goal research setting achieve accounting behavior multiple meet make constraints differing ability particularly association set single conflicting promotes and/or