Advancing the assessment of automated deception detection systems: Incorporating base rate and cost into system evaluation

Twitchell, Douglas P.; Fuller, Christie M.

doi:10.1111/isj.12231

Cited by 9 publications

(8 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The prevalence of deception, social engineering and other forms of deceptive communication online shows no signs of abating anytime soon (Das et al , 2022; Jones et al , 2021; Van der Walt and Eloff, 2019; Xu and Rajivan, 2023). To reduce the damage that ensues after an individual is victimized, training Internet users to detect and disregard deceptive communication seems to be the only practical solution, at least until automated means of filtering deception are implemented and made widely available (Twitchell and Fuller, 2019; Twyman et al , 2015). If anything, the increased use of interpersonal communicative technologies has most likely exacerbated the problem (Siering, 2019).…”

Section: Discussionmentioning

confidence: 99%

Detecting deception in computer-mediated communication: the role of popularity information across media types

Mirsadikov,

Vedadi,

Marett

2024

ICS

View full text Add to dashboard Cite

Purpose With the widespread use of online communications, users are extremely vulnerable to a myriad of deception attempts. This study aims to extend the literature on deception in computer-mediated communication by investigating whether the manner in which popularity information (PI) is presented and media richness affects users’ judgments. Design/methodology/approach This study developed a randomized, within and 2 × 3 between-subject experimental design. This study analyzed the main effects of PI and media richness on the imitation magnitude of veracity judges and the effect of the interaction between PI and media richness on the imitation magnitude of veracity judges. Findings The manner in which PI is presented to people affects their tendency to imitate others. Media richness also has a main effect; text-only messages resulted in greater imitation magnitude than those viewed in full audiovisual format. The findings showed an interaction effect between PI and media richness. Originality/value The findings of this study contribute to the information systems literature by introducing the notion of herd behavior to judgments of truthfulness and deception. Also, the medium over which PI was presented significantly impacted the magnitude of imitation tendency: PI delivered through text-only medium led to a greater extent of imitation than when delivered in full audiovisual format. This suggests that media richness alters the degree of imitating others’ decisions such that the leaner the medium, the greater the expected extent of imitation.

show abstract

Section: Discussionmentioning

confidence: 99%

Detecting deception in computer-mediated communication: the role of popularity information across media types

Mirsadikov,

Vedadi,

Marett

2024

ICS

View full text Add to dashboard Cite

show abstract

“…To emulate real-world conditions, we also evaluated biomarkers with simulations that took into account the base rate of autistic prevalence in the sample. 55,57 To do this, we used 100 linearly spaced base rates between .001 and 1. For each base rate, 100,000 participants were drawn randomly with replacement from the observed distributions of autistic and control biomarker values in proportion to the target base rate.…”

Section: Evaluating Diagnostic Accuracy Of Eeg Measures As Biomarkersmentioning

confidence: 99%

Lack of univariate, clinically-relevant biomarkers of autism in resting state EEG: a study of 776 participants

Dede,

Xiao,

Vaci

et al. 2023

Preprint

View full text Add to dashboard Cite

Neurodevelopmental disorders are difficult to diagnose, requiring expert clinicians and subjective judgements. Consequently, there has been interest in finding quantitative biomarkers for some disorders using resting state electroencephalogram (EEG) data. Here, we focus on resting state EEG biomarkers of autism. Although many previous reports have pointed to EEG-based differences between autistic and neurotypical participants, results have often failed to replicate and sample sizes have typically been small. Taking a big-data, open-science approach, we combined data from five studies to create a large sample of autistic and neurotypical individuals (n = 776) and used high-power computing to extract 942 variables from each participant’s data. Using a systematic, preregistered analysis pipeline, we failed to identify even a single EEG-based variable that could serve as a practically useful biomarker of autism clinical diagnosis. Our results highlight that a biomarker for autism drawn from resting state EEG data is an elusive construct that may not exist.

show abstract

“…To illustrate the importance of contextual information, both theoretical and experimental evidence was provided by [67]. Real world data…”

Section: Primary Datamentioning

confidence: 99%

Systemic Acquired Critique of Credit Card Deception Exposure through Machine Learning

Dantas

Firdaus²,

Jaleel³

et al. 2022

Journal of Open Innovation: Technology, Market, and Complexity

View full text Add to dashboard Cite

A wide range of recent studies are focusing on current issues of financial fraud, especially concerning cybercrimes. The reason behind this is even with improved security, a great amount of money loss occurs every year due to credit card fraud. In recent days, ATM fraud has decreased, while credit card fraud has increased. This study examines articles from five foremost databases. The literature review is designed using extraction by database, keywords, year, articles, authors, and performance measures based on data used in previous research, future research directions and purpose of the article. This study identifies the crucial gaps which ultimately allow research opportunities in this fraud detection process by utilizing knowledge from the machine learning domain. Our findings prove that this research area has become most dominant in the last ten years. We accessed both supervised and unsupervised machine learning techniques to detect cybercrime and management techniques which provide evidence for the effectiveness of machine learning techniques to control cybercrime in the credit card industry. Results indicated that there is room for further research to obtain better results than existing ones on the basis of both quantitative and qualitative research analysis.

show abstract

Advancing the assessment of automated deception detection systems: Incorporating base rate and cost into system evaluation

Cited by 9 publications

References 59 publications

Detecting deception in computer-mediated communication: the role of popularity information across media types

Detecting deception in computer-mediated communication: the role of popularity information across media types

Lack of univariate, clinically-relevant biomarkers of autism in resting state EEG: a study of 776 participants

Systemic Acquired Critique of Credit Card Deception Exposure through Machine Learning

Contact Info

Product

Resources

About