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

Zhou, Lina; Burgoon, Judee K.; Twitchell, Douglas P.; Qin, Tiantian; Nunamaker, Jay F.

doi:10.1080/07421222.2004.11045779

Cited by 207 publications

(213 citation statements)

References 36 publications

(53 reference statements)

Supporting

Mentioning

177

Contrasting

Unclassified

Order By: Relevance

“…The literature also associates deception with decreased usage of first-person singular pronouns, an effect attributed to psychological distancing, whereby deceivers talk less about themselves due either to a lack of personal experience, or to detach themselves from the lie (Newman et al, 2003;Zhou et al, 2004;Knapp and Comaden, 1979). However, according to our findings, we find the opposite to hold.…”

Section: First-person Singular Pronounscontrasting

confidence: 57%

Towards a General Rule for Identifying Deceptive Opinion Spam

Li¹,

Ott²,

Cardie³

et al. 2014

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

271

223

View full text Add to dashboard Cite

Consumers' purchase decisions are increasingly influenced by user-generated online reviews. Accordingly, there has been growing concern about the potential for posting deceptive opinion spamfictitious reviews that have been deliberately written to sound authentic, to deceive the reader. In this paper, we explore generalized approaches for identifying online deceptive opinion spam based on a new gold standard dataset, which is comprised of data from three different domains (i.e. Hotel, Restaurant, Doctor), each of which contains three types of reviews, i.e. customer generated truthful reviews, Turker generated deceptive reviews and employee (domain-expert) generated deceptive reviews. Our approach tries to capture the general difference of language usage between deceptive and truthful reviews, which we hope will help customers when making purchase decisions and review portal operators, such as TripAdvisor or Yelp, investigate possible fraudulent activity on their sites. 1

show abstract

Section: First-person Singular Pronounscontrasting

confidence: 57%

Towards a General Rule for Identifying Deceptive Opinion Spam

Li¹,

Ott²,

Cardie³

et al. 2014

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

271

223

View full text Add to dashboard Cite

show abstract

“…However, this comes at the cost of a user's evaluation time, and can increase the number of false alarms (Biros et al, 2002). Proposals have also been made for automated deception detection (Zhou et al, 2004). Recently proposed credibility assessment systems are capable of detecting individuals' purposely hidden information (Twyman et al, 2014).…”

Section: Deception and Technologymentioning

confidence: 99%

Manipulative Imputation in Distributed Decision Support Settings: The Implications of Information Asymmetry and Aggregation Complexity

Malekovic¹,

Sutanto

Goutas

2016

Decision Support Systems

View full text Add to dashboard Cite

According to earlier research, distributed decision support structures are susceptible to deception. We complement the existing works by analyzing group members' attempts to manipulate group decisions supported by distributed communications. Experimentally, we manipulated two systemic features of a distributed support structure: the members' information asymmetry and decision rule complexity. Both of these features refer to structural properties of aggregated information exchange. We confirmed several hypothesized effects: An increase in the information asymmetry in the aggregation of information increases the incidence of the members' manipulative tendency. It also increases the effectiveness of the members' manipulative imputations. However, the complexity of a decision rule negatively moderates both of these effects. We point out the theoretical relevance and managerial implications of our findings. We conclude that managing team members' information asymmetry and complexity of issues under their practical consideration may result in valuable disclosures.

show abstract

“…On the other hand, deceivers may need more words to create a plausible and convincing story. For example, Zhou et al (2004a) and Zhou et al (2004b) showed that deceptive e-mail messages contained a higher number of words and sentences. Similarly, Burgoon et al (2003) demonstrated, using mock scenes, that deceivers used longer messages when communicating about their experience.…”

Section: Descriptive Featuresmentioning

confidence: 99%

“…This difference can be explained by the fact that the writer of an annual report has more time to create the deceptive text than a speaker on a call. Zhou et al (2004b) found that lexical diversity is an important feature in their machine learning algorithm for detecting deception in computer-mediated communication. The deceptive messages showed less lexical diversity (Zhou et al, 2004a).…”

Section: Descriptive Featuresmentioning

confidence: 99%

“…In computer-mediated communication, the deceptive messages contained less punctuation, fewer long sentences and fewer syllables per word (Zhou et al, 2003). Therefore, Zhou et al (2004b) used these features in their machine learning model to detect such messages. The experiments with deceptive messages of Burgoon et al (2003) showed a similar result.…”

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

Cited by 207 publications

References 36 publications

Towards a General Rule for Identifying Deceptive Opinion Spam

Towards a General Rule for Identifying Deceptive Opinion Spam

Manipulative Imputation in Distributed Decision Support Settings: The Implications of Information Asymmetry and Aggregation Complexity

Text mining to detect indications of fraud in annual reports worldwide

Contact Info

Product

Resources

About