Discriminative Models for Predicting Deception Strategies

Abstract

Although a large body of work has previously investigated various cues predicting deceptive communications, especially as demonstrated through written and spoken language (e.g., [30]), little has been done to explore predicting kinds of deception. We present novel work to evaluate the use of textual cues to discriminate between deception strategies (such as exaggeration or falsification), concentrating on intentionally untruthful statements meant to persuade in a social media context. We conduct human subjects experimentation wherein subjects were engaged in a conversational task and then asked to label the kind(s) of deception they employed for each deceptive statement made. We then develop discriminative models to understand the difficulty between choosing between one and several strategies. We evaluate the models using precision and recall for strategy prediction among 4 deception strategies based on the most relevant psycholinguistic, structural, and data-driven cues. Our single strategy model results demonstrate as much as a 58% increase over baseline (random chance) accuracy and we also find that it is more difficult to predict certain kinds of deception than others.

A Framework for Detecting Intentions of Criminal Acts in Social Media: A Case Study on Twitter

“…Appling et al [63] studied the use of linguistic techniques and discriminatory model to detect fraud in texts. One of the linguistic techniques used was the amount of words used in a given text to affirm something.…”

Using Data Analytics to Filter Insincere Posts from Online Social Networks A Case Study: Quora Insincere Questions

“…Many authors looked into cues for deception in OSN posts [14][15][16][17]. In OSNs, cues of deceptions that are available for face to face communications (e.g., eye contact, gaze aversion, shrugs, amplitude, etc.)…”

“…Appling et al [16] described different types of deception strategies, including Falsification, exaggeration, omission, and misleading. Deceptions can also be categorized based on strategies and models and also based on intent to deceive [17].…”

Detecting deceptive reviews using Argumentation

“…However, the authenticity and truthfulness of these reviews is not guaranteed and content communities, review and news websites are susceptible to deceptive content. Different deception strategies exist: falsification (contradictions/ lies), exaggeration (superlative information), omission (hiding information) and misleading information (irrelevant information/ topic changes) [1]. Opinions expressed in online forums or e-commerce websites attract and influence potential customers.…”

“…Although they are not considered to be malware, deceptive reviews can pose risks for security and privacy by persuading potential customers to buy a company's product/ service given positive reviews or discouraging customers from purchasing when faced with negative reviews. Deception technology has been acknowledged as an emerging security technology 1 . Whilst threat deception can be used in some settings, in this paper we focus on a very specific type of deception, namely deceptive reviews.…”