Situational Contingencies in Susceptibility of Social Media to Phishing: A Temptation and Restraint Model

2023 | Journal of Management Information Systems | Citations: 0

Authors: Qahri-Saremi, Hamed; Turel, Ofir

Abstract: User susceptibility to phishing messages on social media is a growing informatio ... Expand

Abstract: User susceptibility to phishing messages on social media is a growing information security concern. Contingency factors that can influence this susceptibility and the theoretical mechanisms through which they operate need more scholarly attention. To bridge this gap, we present a temptation and restraint (TR) model (a specific manifestation of the dual–system theory) of social media phishing susceptibility, which explains it as an outcome of a struggle between users' temptation toward engaging with a social media phishing message and their cognitive and behavioral restraint against it. The balance in this struggle is a function of various situational contingencies. First, via a Delphi study, we identify four key situational contingency factors in the context of social media that can influence this balance: (1) poor sleep quality, (2) social media ostracism, (3) source likability, and (4) fear appeals. Next, via five randomized controlled experiments using an ostensible social media paradigm with social media users, we show that the TR model explains (a) why and how users engage with social media phishing messages, and (b) when users are more or less susceptible to it based on key situational contingency factors. Our findings offer a nuanced perspective on social media phishing susceptibility, elucidate the fundamental roles of situational contingencies in the genesis of social media phishing victimization, and delineate important directions for future research in this area Collapse

Topics: social media phishing armed conflict decision making user behavior

Methods: experiment regression analysis method delphi study analysis of covariance mixed method

Theories: big five model dual-system theory

Phishing Susceptibility in Context: A Multilevel Information Processing Perspective on Deception Detection

2023 | Management Information Systems Quarterly | Citations: 0

Authors: Wright, Ryan; Johnson, Steven; Kitchens, Brent

Abstract: Despite widespread awareness of risks, significant investments in cybersecurity ... Expand

Abstract: Despite widespread awareness of risks, significant investments in cybersecurity protection, and substantial economic incentives to avoid security breaches, organizations remain vulnerable to phishing attacks. Phishing research has informed effective practical interventions to address phishing susceptibility that emphasize the importance of broadly applicable IT security knowledge. Yet employees still frequently fall victim to phishing attempts. To help understand why, we conceptualize phishing susceptibility as the failure to differentiate between deceptive and legitimate information processing requests that occur within the context of an employee's typical job responsibilities. We apply this contextual lens to identify characteristics of knowledge workers' organizational task and social context that may enhance or diminish performance in detecting deception in phishing email attempts. To test our hypotheses, we conducted a study in which employees of the finance division of a large university encountered simulated email-based phishing attempts as part of their normal work routine. We found evidence supporting our hypotheses that an individual's susceptibility to phishing attacks is influenced by their position in the knowledge flows of the organization and by the impact of workgroup responsibilities on their cognitive processing. We contend that phishing susceptibility is not merely a matter of IT security knowledge but is also influenced by contextualized, multilevel influences on information processing. As phishing attacks are increasingly targeted to specific organizational settings, it is even more important to incorporate this contextualized information processing view of phishing susceptibility. Collapse

Topics: phishing electronic mail system support IT security email phishing

Methods: theory development literature study simulation logistic regression survey

Analysis of third-party request structures to detect fraudulent websites

2022 | Decision Support Systems | Citations: 0

Authors: Gopal, Ram D.; Hojati, Afrouz; Patterson, Raymond A.

Abstract: Third-party websites or applications are the key entities in the web eco-system ... Expand

Abstract: Third-party websites or applications are the key entities in the web eco-system that enable websites to function and offer services. Almost every organization today uses dozens of websites and sub-domains. Each provides essential functions and typically uses dozens of third-parties to produce its capabilities. With the growing problem of illegitimate websites, such as those peddling fake news and selling counterfeit products, the detection of fraudulent websites becomes more and more crucial. While the conventional method of fraudulent website detection mostly relies on the content-based analysis of websites, the method of this study uses third-party request structure features and attributes of third-parties engaged in the structure to predict legitimate and fraudulent websites. This method can be used on a real-time basis to complement current detection methods. Moreover, our approach is not limited to a specific category of websites. In other words, unlike previous studies, our approach is able to increase the likelihood of detecting all kinds of fake and fraudulent websites. The results of this study are largely robust across different predictive models. Collapse

Topics: website phishing website electronic commerce advertising management fake news

Methods: machine learning classification method data modeling ensemble learning computational algorithm

Readmission Prediction for Patients with Heterogeneous Medical History: A Trajectory-Based Deep Learning Approach

2022 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Xie, Jiaheng; Zhang, Bin; Ma, Jian; Zeng, Daniel; Lo-Ciganic, Jenny

Abstract: Hospital readmission refers to the situation where a patient is re-hospitalized ... Expand

Abstract: Hospital readmission refers to the situation where a patient is re-hospitalized with the same primary diagnosis within a specific time interval after discharge. Hospital readmission causes $26 billion preventable expenses to the U.S. health systems annually and often indicates suboptimal patient care. To alleviate those severe financial and health consequences, it is crucial to proactively predict patients’ readmission risk. Such prediction is challenging because the evolution of patients’ medical history is dynamic and complex. The state-of-the-art studies apply statistical models which use static predictors in a period, failing to consider patients’ heterogeneous medical history. Our approach – Trajectory-BAsed DEep Learning (TADEL) – is motivated to tackle the deficiencies of the existing approaches by capturing dynamic medical history. We evaluate TADEL on a five-year national Medicare claims dataset including 3.6 million patients per year over all hospitals in the United States, reaching an F1 score of 87.3% and an AUC of 88.4%. Our approach significantly outperforms all the state-of-the-art methods. Our findings suggest that health status factors and insurance coverage are important predictors for readmission. This study contributes to IS literature and analytical methodology by formulating the trajectory-based readmission prediction problem and developing a novel deep-learning-based readmission risk prediction framework. From a health IT perspective, this research delivers implementable methods to assess patients’ readmission risk and take early interventions to avoid potential negative consequences. Collapse

Topics: health information system healthcare data online learning system crowdsourcing crowdfunding

Methods: machine learning deep learning logistic regression support vector machine experimental group

The Phishing Funnel Model: A Design Artifact to Predict User Susceptibility to Phishing Websites

2021 | Information Systems Research | Citations: 17

Authors: Abbasi, Ahmed; Dobolyi, David; Vance, Anthony; Zahedi, Fatemeh Mariam

Abstract: Phishing is a significant security concern for organizations, threatening employ ... Expand

Abstract: Phishing is a significant security concern for organizations, threatening employees and members of the public. Phishing threats against employees can lead to severe security incidents, whereas those against the public can undermine trust, satisfaction, and brand equity. At the root of the problem is the inability of Internet users to identify phishing attacks even when using anti-phishing tools. We propose the phishing funnel model (PFM), a design artifact for predicting user susceptibility to phishing websites. PFM incorporates user, threat, and tool-related factors to predict actions during four key stages of the phishing process: visit, browse, consider legitimate, and intention to transact. We used a support vector ordinal regression with a custom kernel encompassing a cumulative-link mixed model for representing users’ decisions across funnel stages. We evaluated the efficacy of PFM in a 12-month longitudinal field experiment in two organizations involving 1,278 employees and 49,373 phishing interactions. PFM significantly outperformed competing models/methods by 8%–52% in area under the curve, correctly predicting visits to high-severity threats 96% of the time—a result 10% higher than the nearest competitor. A follow-up three-month field study revealed that employees using PFM were significantly less likely to interact with phishing threats relative to comparison models and baseline warnings. Furthermore, a cost-benefit analysis showed that interventions guided by PFM resulted in phishing-related cost reductions of nearly $1,900 per employee more than comparison prediction methods. These results indicate strong external validity for PFM. Our findings have important implications for practice by demonstrating (1) the effectiveness of predicting user susceptibility to phishing as a real-time protection strategy, (2) the value of modeling each stage of the phishing process together, rather than focusing on a single user action, and (3) the considerable impact of anti-phishing tool and threat-related factors on susceptibility to phishing. Collapse

Topics: phishing electronic mail phishing website website cost benefit analysis

Methods: survey experiment field experiment machine learning longitudinal research

Theories: technology acceptance model protection motivation theory

Trust calibration of automated security IT artifacts: A multi-domain study of phishing-website detection tools

2021 | Information & Management | Citations: 0

Authors: Chen, Yan; Zahedi, Fatemeh Mariam; Abbasi, Ahmed; Dobolyi, David

Abstract: Phishing websites become a critical cybersecurity threat affecting individuals a ... Expand

Abstract: Phishing websites become a critical cybersecurity threat affecting individuals and organizations. Phishingwebsite detection tools are designed to protect users against such sites. Nevertheless, detection tools face serious user trust and suboptimal performance issues which require trust calibration to align trust with the tool’s capabilities. We employ the theoretical framework of automation trust and reliance as a kernel theory to develop the trust calibration model for phishing-website detection tools. We test the model using a controlled lab experiment. The results of our analysis show that users’ trust in detection tools can be calibrated by trust cali­ brators. Moreover, users’ calibrated trust has significant consequences, including users’ tool reliance, use, and performance against phishing websites. Collapse

Topics: phishing website website usage intention online banking IT security

Methods: experiment laboratory experiment structural equation modeling kernel theory factor analysis

Eyes wide open: The role of situational information security awareness for security-related behaviour

2021 | Information Systems Journal | Citations: 0

Authors: Jaeger, Lennart; Eckhardt, Andreas

Abstract: Most contemporary studies on information security focus on largely static phenom ... Expand

Abstract: Most contemporary studies on information security focus on largely static phenomena in examining security-related behaviours. We take a more dynamic, situational and interactionist approach that proposes that security-related behaviours result from an interaction between the person and the perception of a threatening situation. We derive and define situational information security awareness based on situation awareness literature, and examine how individual-level (innate traits, experience) and system-level factors (design variations, warning signal) influence awareness, and how it influences subsequent threat and coping appraisals, and ultimately security-related behaviours in a multi-method phishing experiment including eye tracking and survey components with 107 employees. The results underscore the importance of situational information security awareness and show that past experience with phishing and a security warning increase awareness, while phishing emails' contextual relevance and misplaced salience decrease awareness. Situational information security awareness, in turn, increases perceived threat and perceived coping efficacy and, ultimately, actual behavioural responses to phishing attacks. Collapse

Topics: cybersecurity awareness phishing electronic mail email phishing IT security threat

Methods: oculometry experiment survey partial least squares regression statistical power analysis

Theories: protection motivation theory big five model

Design Principles for Robust Fraud Detection: The Case of Stock Market Manipulations

2021 | Journal of the Association for Information Systems | Citations: 11

Authors: Siering, Michael; Muntermann, Jan; Grčar, Miha

Abstract: We address the challenge of building an automated fraud detection system with ro ... Expand

Abstract: We address the challenge of building an automated fraud detection system with robust classifiers that mitigate countermeasures from fraudsters in the field of information-based securities fraud. Our work involves developing design principles for robust fraud detection systems and presenting corresponding design features. We adopt an instrumentalist perspective that relies on theory-based linguistic features and ensemble learning concepts as justificatory knowledge for building robust classifiers. We perform a naive evaluation that assesses the classifiers’ performance to identify suspicious stock recommendations, and a robustness evaluation with a simulation that demonstrates a response to fraudster countermeasures. The results indicate that the use of theory-based linguistic features and ensemble learning can significantly increase the robustness of classifiers and contribute to the effectiveness of robust fraud detection. We discuss implications for supervisory authorities, industry, and individual users. Collapse

Topics: fraud detection marketing management problem solving advertising management evaluation criteria

Methods: design principle design requirement ensemble learning machine learning kernel theory

Theories: marketing theory

An examination of the effect of recent phishing encounters on phishing susceptibility

2020 | Decision Support Systems | Citations: 24

Authors: Chen, Rui; Gaia, Joana; Rao, H. Raghav

Abstract: This paper examines online users' perceived susceptibility to phishing attacks. ... Expand

Abstract: This paper examines online users' perceived susceptibility to phishing attacks. We posit that an individual's phishing susceptibility may be shaped by recent phishing encounters and, more importantly, that the effect of new experience on susceptibility will be heterogeneous among users. To facilitate our investigation, we focus on both the process and outcome of phishing detection. Survey data from college students confirms that one's susceptibility is affected by detection process difficulty and detection outcome failures in the recent phishing encounter. Results also reveal the importance of personal attributes, such as past success in phishing detection and phishing desensitization, in regulating the effects of a recent phishing encounter. Finally, results show the relationship between detection process difficulty and outcome failures, in addition to confirming antecedents to the two detection components. Our research generates new knowledge that contributes to the phishing literature and it also sheds new insights that inform practitioners, although the use of college students limits the generalizability of the current findings. Collapse

Topics: phishing electronic mail decision making internet technology email phishing

Methods: survey descriptive statistic partial least squares regression machine learning structural equation modeling

Theories: expectation confirmation theory

Real or Not? Identifying Untrustworthy News Websites Using Third-party Partnerships

2020 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Gopal, Ram D.; Hidaji, Hooman; Kutlu, Sule Nur; Patterson, Raymond A.; Rolland, Erik; Zhdanov, Dmitry

Abstract: Untrustworthy content such as fake news and clickbait have become a pervasive pr ... Expand

Abstract: Untrustworthy content such as fake news and clickbait have become a pervasive problem on the Internet, causing significant socio-political problems around the world. Identifying untrustworthy content is a crucial step in countering them. The current best practices for identification involve content analysis and arduous fact-checking of the content. To complement content analysis, we propose examining websites’ third-parties to identify their trustworthiness. Websites utilize third-parties, also known as their digital supply chains, to create and present content and help the website function. Third-parties are an important indication of a website's business model. Similar websites exhibit similarities in the third-parties they use. Using this perspective, we use machine learning and heuristic methods to discern similarities and dissimilarities in third-party usage, which we use to predict trustworthiness of websites. We demonstrate the effectiveness and robustness of our approach in predicting trustworthiness of websites from a database of News, Fake News, and Clickbait websites. Our approach can be easily and cost-effectively implemented to reinforce current identification methods. Collapse

Topics: website fake news supply chain management business model ease of use

Methods: support vector machine machine learning k-means clustering classification method ensemble learning