Deception detection for the tangled web

Abstract

Deception is a reasonably common part of daily life that society sometimes demonstrates a degree of acceptance of, and occasionally people are very willing to be deceived. But can a computer identify deception and distinguish it from that which is not deceptive?We explore deception in various guises, differentiating it from lies, and highlighting the influence of medium and message in both deception and its detection. Our investigations to date have uncovered disagreements relating to the measurements of such cues, and variations in interpretations, as could be problematic in building a deception detection system.

Digital forensics supported by machine learning for the detection of online sexual predatory chats

“…[50] cosine similarity metric Dictionary of terms 0.0498 [50] cosine similarity metric multinomial Naive Bayes 0.0354 [52] pattern matching similarity 0.48 [51] rules JRIP 0.47 [53] n-grams k-NN 0.70 we sequentially merged these segments based on the combinations resulting in the highest prediction probability. We do this until the prediction probability is the highest it can be or there are only two segments left.…”

“…Vartapetiance and Gillam [52] obtained an F 1 score of 0.48 by manually constructing rules. Better performance is obtained by Kang et al [53] with an F 1 score of 0.7 using K-nearest Neighbours (KNN).…”

Detecting advance fee fraud emails using self-referential pronouns: A preliminary analysis

“…These first person singular pronouns have been associated with deceptive communication (Gupta and Skillicorn (2006); Newman, Pennebaker, Berry, and Richards (2003)). Their use declares an individual's ownership of a statement and is tantamount to a projection of honesty (Vartapetiance & Gillam, 2012).…”

Truth and deception at the rhetorical structure level

“…5 Rubin and Chen's (2012) information manipulation classification theory conceptualizes and differentiates broad types of information manipulation (including falsification, exaggeration, concealment, and hoax) by 12 salient factors (such as intentionality, accuracy, and social acceptability). 6 In retrospect, some researchers question whether there is "a suitably computable basis" for a mechanism of identifying deception and building a deception detection system based on lexicosemantic deception cues because of disagreements in measuring and interpreting such cues (Vartapetiance & Gillam, 2012). 7 Using a corpus of criminal statements, police interrogations, and legal testimonies, their regression and tree-based classification automatic tagger performs at average 69% recall and 85% precision rates compared with the performance of human taggers on the same subset (Bachenko et al, 2008). this study focuses on one main question: What is the impact of the relations among discourse constituent parts on the discourse composition of deceptive and truthful messages?…”

“…In retrospect, some researchers question whether there is “a suitably computable basis” for a mechanism of identifying deception and building a deception detection system based on lexicosemantic deception cues because of disagreements in measuring and interpreting such cues (Vartapetiance & Gillam, ).…”