A critical analysis of the state-of-the-art on automated detection of deceptive behavior in social media

Recently, a large body of research has been devoted to examine the user behavioral patterns and the business implications of social media. However, relatively little research has been conducted regarding users' deceptive activities in social media; these deceptive activities may hinder the effective application of the data collected from social media to perform e-marketing and initiate business transformation in general. One of the main contributions of this paper is the critical analysis of the possible forms of deceptive behavior in social media and the state-of-the-art technologies for automated deception detection in social media. Based on the proposed taxonomy of major deception types, the assumptions, advantages, and disadvantages of the popular deception detection methods are analyzed. Our critical analysis shows that deceptive behavior may evolve over time, and so making it difficult for the existing methods to effectively detect social media spam. Accordingly, another main contribution of this paper is the design and development of a generic framework to combat dynamic deceptive activities in social media. The managerial implication of our research is that business managers or marketers will develop better insights about the possible deceptive behavior in social media before they tap into social media to collect and generate market intelligence. Moreover, they can apply the proposed adaptive deception detection framework to more effectively combat the ever increasing and evolving deceptive activities in social media.