Search Results (109)

Sextortion has rapidly expanded into a global cyber-enabled crime that leverages anonymous digital communication and decentralized payment systems. This study examines the financial infrastructures underlying contemporary sextortion by conducting a two-phase analysis of 87 confirmed cases involving cryptocurrency payments. Using blockchain forensic tools and open-source intelligence, the research traces fund movements across perpetrator-controlled wallets, identifies laundering techniques such as mixers, peel-chain transfers, and exchange-based cash-outs, and links these behaviors to narrative patterns within victim reports. The results reveal a dual-tier ecosystem in which mass-produced, multilingual extortion scripts coexist with divergent laundering typologies that differentiate lower-value, high-volume scams from more organized and higher-yield operations. By integrating qualitative and quantitative evidence, this study provides a forensic framework for detecting illicit cryptocurrency activity, improving threat classification, and strengthening investigative and regulatory responses to sextortion and related crypto-enabled interpersonal crimes. Full article

Real-time telecom scam detection is difficult without cloud AI, particularly for privacy-sensitive, low-resource environments. We developed a lightweight, offline voice scam detector using on-device audio segmentation, automatic speech recognition (ASR), and semantic similarity. Four detection strategies were implemented. We used Whisper ASR and DeepSeek to process 5 s speech chunks. An analysis of 120 synthetic and paraphrased Mandarin phone call dialogues reveals the A4 voting strategy’s superior performance in optimizing early detection and minimizing false positives, achieving an F1 score of 0.90, a 2.5% false positive rate, and a mean response time of under 4 s. The system is deployable on ESP32 for offline mobile inference. The proposed architecture provides a robust and scalable defense against threats targeting vulnerable user groups, such as older adults. It introduces a new method for real-time voice threat mitigation on devices through interpretable segment-level semantic analysis. Full article

Introduction: Several studies have been conducted in the field of education for older adults, with an emphasis on teaching and learning processes related to the use of digital technologies. Among the relevant aspects to be considered in this context is the cognitive vulnerability of this age group in terms of digital security. Objective: The aim of this study was to analyze the relationship between cognitive aspects of older adults and their ability to identify digital risks, before and after participating in an educational intervention, as well as the effect of the intervention on cognition in this age group. Methodology: Analyses were conducted according to the educational intervention and control groups, further stratified by digital risk (SJT) and dementia risk, according to the ACE-R test. The Mann–Whitney test was used to identify possible differences in the likelihood of falling for digital scams, considering the dimensions generated by the simulations (SJT). Results: Overall, the educational intervention was effective for the media education dimension (delta −0.5), regardless of dementia risk. More specifically, a particular effect was observed in the post-intervention stage. Conclusions: The educational intervention was able to promote cognitive gains and reduce digital risks among older adults, particularly in the identification of misinformation, underscoring the importance of continuous and adapted programs to promote digital security in this population. Full article

As stablecoins become increasingly prevalent in financial crimes, their usage for illicit activities has reached a scale of USD 51.3 billion. Detecting phishing activities within stablecoin transactions has emerged as a critical challenge in blockchain security. Currently, existing detection methods predominantly target mainstream cryptocurrencies like Ethereum and lack specialized models tailored to the unique transaction patterns of stablecoin networks. This paper introduces a deep learning framework, BERTSC, based on multi-modal fusion. The model integrates three core modules graph convolutional networks (GCNs), BERT semantic encoders, and soft prompt encoders to identify malicious accounts. The GCN constructs directed multi-graph representations of account interactions, incorporating multi-dimensional edge features; the BERT encoder transforms discrete transaction attributes into semantically rich continuous vector representations; the soft prompt encoder maps account interaction features into learnable prompt vectors. An innovative three-way gated dynamic fusion mechanism optimally combines the information from these sources. The fused features are then classified to predict phishing account labels, facilitating the detection of phishing scams in stablecoin transaction datasets. Experimental results on large-scale stablecoin datasets demonstrate that BERTSC outperforms baseline models, achieving improvements of 4.96%, 3.60%, and 4.23% in Precision, Recall, and F1-score, respectively. Ablation studies validate the effectiveness of each module and confirm the necessity and superiority of the three-way gating fusion mechanism. This research offers a novel technical approach for phishing detection within blockchain stablecoin ecosystems. Full article

Wangiri fraud is a pervasive telecommunications scam that exploits missed calls to lure victims into dialing premium-rate numbers, resulting in significant financial losses for operators and consumers. This paper presents a comprehensive machine learning framework for detecting Wangiri fraud in highly imbalanced and unlabeled Call Detail Record (CDR) datasets. We introduce a novel unsupervised labeling approach using domain-driven heuristics, coupled with advanced feature engineering to capture temporal, geographic, and behavioral patterns indicative of fraud. To address severe class imbalance, we evaluate multiple sampling strategies like the Synthetic Minority Over-sampling Technique (SMOTE) and undersampling, and also compare the performance of Logistic Regression, Decision Trees, Random Forest, XGBoost, and Multi-Layer Perceptron (MLP). Our results demonstrate that ensemble methods, particularly Random Forest and XGBoost, achieve near-perfect accuracy (e.g., Receiver Operating Characteristic Area Under the Curve (ROC-AUC) $>0.99$) on balanced data while maintaining interpretability. The proposed pipeline offers a scalable and practical solution for real-time fraud detection, providing telecom operators with an effective tool to mitigate Wangiri fraud risks. Full article

The Cyber Kill Chain (CKC) is a prevalent concept in cyber defense; nevertheless, its emphasis on post-reconnaissance phases limits the capacity to foresee attacker activities outside the organizational boundary. This study introduces and empirically substantiates a pre-chain phase, referred to as contextual anticipation, which broadens the temporal framework of the CKC by methodically identifying subtle yet actionable signals prior to reconnaissance. The methodology combines the STEMPLES+ framework for socio-technical scanning with General Morphological Analysis (GMA), generating internally coherent scenarios that are translated into Indicators of Threats (IOT). These indicators connect contextual triggers to threshold-based monitoring activities and established courses of action, forming a reproducible and auditable relationship between foresight analysis and operational defense. The application of three illustrative cases—a banking merger, the distribution of a phishing kit in underground marketplaces, and wartime contribution scams—illustrated that contextual anticipation consistently provided quantifiable lead-time benefits varying from several days to six weeks. This proactive stance enabled measures such as registrar takedowns, targeted awareness campaigns, and anticipatory monitoring before distribution and exploitation. By formalizing CKC-0 as an integrated socio-technical phase, the research enhances cybersecurity practice by demonstrating how diffuse contextual drivers can be converted into organized, actionable mechanisms for proactive resilience. Full article

Transaction Fraud, a type of financial operational risk, remains a major threat to financial sectors and continuously imposes devastating financial impacts. This study comprehensively reviews 41 cutting-edge publications on financial transaction fraud detection using Machine Learning from January 2018 to October 2025. We establish a taxonomy to categorize the selected work into four themes: Traditional Machine Learning, Deep Learning, Ensemble Method, and Hybrid Method. Each theme is evaluated in-depth, from strengths to weaknesses. Ensemble exhibits better performance over other methods with a recall of 92.7%, a precision of 96% and an F1-score of 92.66% on average, while Traditional ML ranks last in terms of average F1-score. Preprocessing strategies, like data balancing, can enhance performance, while feature engineering requires careful evaluation before implementation. Significantly, we assess financial implications, suggesting it is essential to integrate financial metric design, feature explanation, time series patterns, and data privacy considerations into financial fraud detection—a focus that aligns with risk management frameworks and regulations. By revealing current research gaps and suggesting future directions, our study provides practical guidance for researchers and practitioners to advance financial fraud detection strategies within a highly intricate financial ecosystem. Full article

As digital literacy becomes central to cybercrime prevention, we examine how adults of different ages engage with online learning, moving beyond age alone to consider additional drivers of preference. We analyzed a nationally representative U.S. adult sample (N = 1113; Nov 2024). Ordinal logistic regressions assessed associations between preferences for cybersecurity education and age, education, income, subjective well-being (SWB), and high-speed internet access. Interaction terms (e.g., age × internet access) were tested but not retained. Preferences declined with age across most tools, with the sharpest drop being for highly interactive or novel formats (VR/AR, gamification). Actor-based, non-interactive videos showed no age advantage. Education displayed selective positive links, especially for interactive features, while income was largely unrelated. SWB was a broadly enabling correlate, often with nonlinear patterns, and reliable high-speed internet was consistently aligned with stronger preferences. Overall, the model fit was moderate. Effective cybersecurity education should not rely on age-based assumptions. Designing offerings that emphasize clear purpose and ease of use, pair reliable broadband with skills supports, and account for learners’ well-being can improve engagement and potential scam resilience across age groups. Full article

The landscape of cybercrime has undergone significant transformations over the past decade. Present-day threats include AI-generated attacks, deep fakes, 5G network vulnerabilities, cryptojacking, and supply chain attacks, among others. To remain resilient against contemporary threats, it is essential to examine historical data to gain insights that can inform cybersecurity strategies, policy decisions, and public awareness campaigns. This paper presents a comprehensive analysis of the evolution of cyber trends in state-sponsored attacks over the past 20 years, based on the council on foreign relations state-sponsored cyber operations (2005–present). The study explores the key trends, patterns, and demographic shifts in cybercrime victims, the evolution of complaints and losses, and the most prevalent cyber threats over the years. It also investigates the geographical distribution, the gender disparity in victimization, the temporal peaks of specific scams, and the most frequently reported internet crimes. The findings reveal a traditional cyber landscape, with cyber threats becoming more sophisticated and monetized. Finally, the article proposes areas for further exploration through a comprehensive analysis. It provides a detailed chronicle of the trajectory of cybercrimes, offering insights into its past, present, and future. Full article

In real-world dairy farming environments, object recognition models often suffer from missed or false detections due to complex backgrounds and cow occlusions. In response to these issues, this paper proposes FSCA-YOLO, a multi-object cow behavior recognition model based on an improved YOLOv11 framework. First, the FEM-SCAM module is introduced along with the CoordAtt mechanism to enable the model to better focus on effective behavioral features of cows while suppressing irrelevant background information. Second, a small object detection head is added to enhance the model’s ability to recognize cow behaviors occurring at the distant regions of the camera’s field of view. Finally, the original loss function is replaced with the SIoU loss function to improve recognition accuracy and accelerate model convergence. Experimental results show that compared with mainstream object detection models, the improved YOLOv11 in this section demonstrates superior performance in terms of precision, recall, and mean average precision (mAP), achieving 95.7% precision, 92.1% recall, and 94.5% mAP—an improvement of 1.6%, 1.8%, and 2.1%, respectively, over the baseline YOLOv11 model. FSCA-YOLO can accurately extract cow features in real farming environments, providing a reliable vision-based solution for cow behavior recognition. To support specific behavior recognition and in-region counting needs in multi-object cow behavior recognition and tracking systems, OpenCV is integrated with the recognition model, enabling users to meet the diverse behavior identification requirements in groups of cows and improving the model’s adaptability and practical utility. Full article

Bank scams involving the impersonation of law enforcement personnel and financial service providers continue to proliferate across South Africa, leading to substantial economic loss and psychological harm to certain individuals in the country. The persistence of this cyber-enabled fraud indicates a significant lacuna in understanding the systemic vulnerabilities that perpetrators exploit. Specifically, there is a pressing need to examine why these scams remain successful despite existing security measures, identify the key parameters that influence individuals’ susceptibility to deception, and assess the adequacy of current preventive measures. This study navigates these notable concerns using an exploratory case study qualitative research design. Through a non-probabilistic sampling strategy, seven participants were identified to engage in discourse and contributed insights into the subject matter. A nuanced analysis identified an effective monitoring system, heightened public awareness, and stringent penalties as mitigative strategies. Subsequent studies may examine the resultant strategies broadly for wider application. Full article

According to the literature, the advancement of information and communication technology (ICT) has increased individual exposure to scams, turning fraud victimization into a significant concern. While prior research has primarily focused on socio-demographic predictors of fraud victimization, this study adopts a behavioral perspective that is grounded in the Signal Detection Theory (SDT) to investigate the likelihood determinants of individuals becoming fraud victims. Using survey data of 671 Indonesian respondents analyzed with the Partial Least Squares Structural Equation Modeling (PLS-SEM), we explored the roles of vigilance and financial literacy in moderating the relationship between fraud exposure and victimization. Our findings substantiate the notion that higher exposure to fraudulent activity significantly increases the likelihood of victimization. The results also show that vigilance negatively moderates the relationship between fraud exposure and fraud victimization, suggesting that individuals with higher vigilance are better at identifying scams, thereby decreasing their likelihood of becoming fraud victims. Furthermore, financial literacy is positively related to vigilance, indicating that financially literate individuals are more aware of potential scams. However, the predictive power of financial literacy on vigilance is relatively low. Hence, while literacy helps a person sharpen their indicators for detecting fraud, psychological, behavioral, and contextual factors may also affect their vigilance and decision-making. Full article

Dating applications (DAs) are widely used to establish social and sexual connections among men who have sex with men (MSM), particularly in urban areas. In this study, we aimed to examine the associations between DA use and mental health among Thai MSM. An online cross-sectional survey was completed by 442 MSM residing in Bangkok and urban municipalities across all regions of Thailand. Psychological distress (PD) and probable depression were assessed using the General Health Questionnaire (GHQ-12) and the Patient Health Questionnaire (PHQ-9), respectively. Of the participants, 62.7% were current users, with 33.2% experiencing PD and 33.9% having depression. A logistic regression analysis showed that PD was significantly associated with late-night use (AOR = 2.02, 95% CI: 1.08–3.78), matching failure (AOR = 1.95, 95% CI: 1.12–3.38), rejection (AOR = 2.07, 95% CI: 1.18–3.62), and ghosting (AOR = 1.78, 95% CI: 1.02–3.11). Simultaneously, depression was significantly associated with using DAs with the motivation of hooking up (AOR = 2.27, 95% CI: 1.05–4.93), privacy violations (AOR = 2.76, 95% CI: 1.42–5.38), unsolicited sexual images (AOR = 2.04, 95% CI: 1.11–3.74), physical assault (AOR = 2.97, 95% CI: 1.57–5.61), harassment (AOR = 2.54, 95% CI: 1.37–4.70), scams (AOR = 2.59, 95% CI: 1.41–4.77), and extreme disappointment from DA use (AOR = 5.98, 95% CI: 1.84–19.41). These findings highlight how DA usage patterns and negative experiences may contribute to the poorer mental health among MSM in urban areas. Full article

In today’s business environment, emails are essential across all sectors, including finance and academia. There are two main types of emails: ham (legitimate) and spam (unsolicited). Spam wastes consumers’ time and resources and poses risks to sensitive data, with volumes doubling daily. Current spam identification methods, such as Blocklist approaches and content-based techniques, have limitations, highlighting the need for more effective solutions. These constraints call for detailed and more accurate approaches, such as machine learning (ML) and deep learning (DL), for realistic detection of new scams. Emphasis has since been placed on the possibility that ML and DL technologies are present in detecting email spam. In this work, we have succeeded in developing a hybrid deep learning model, where Long Short-Term Memory (LSTM) and the Gated Recurrent Unit (GRU) are applied distinctly to identify spam email. Despite the fact that the other models have been applied independently (CNNs, LSTM, GRU, or ensemble machine learning classifier) in previous studies, the given research has provided a contribution to the existing body of literature since it has managed to combine the advantage of LSTM in capturing the long-term dependency and the effectiveness of GRU in terms of computational efficiency. In this hybridization, we have addressed key issues such as the vanishing gradient problem and outrageous resource consumption that are usually encountered in applying standalone deep learning. Moreover, our proposed model is superior regarding the detection accuracy (90%) and AUC (98.99%). Though Transformer-based models are significantly lighter and can be used in real-time applications, they require extensive computation resources. The proposed work presents a substantive and scalable foundation to spam detection that is technically and practically dissimilar to the familiar approaches due to the powerful preprocessing steps, including particular stop-word removal, TF-IDF vectorization, and model testing on large, real-world size dataset (Enron-Spam). Additionally, delays in the feature comparison technique within the model minimize false positives and false negatives. Full article

Internet scams have become more sophisticated and prevalent in countries such as Canada, the US, the UK, and Australia. Australia has made some progress in effective scam intervention strategies and seen possible growth in public awareness. However, there is a lack of insight into factors associated with profound shame and embarrassment, emotional distress such as anxiety and depression, and trauma and suicide in scam victims. To fill this gap, this perspective paper aimed to provide insight into the factors associated with the negative mental health impacts of internet scams by integrating a narrative literature review with a victim case study detailing a group’s experience of an investment scam in Australia. It found that internet scams cause emotional and social issues like depression, anxiety, trauma, and isolation, mostly prolonged upon substantial loss. The author’s insight into the intensely negative mental health impacts of an investment scam allows for the presentation of a group who struggled to access adequate support and mental health care in their response to insidious organized crime. Better education, resilience-building, and support systems are needed. These shortcomings call for strategies for tailored digital mental health services such as emotionally attuned, trauma-informed digital companionship through human-like artificial intelligence (AI) applications. Full article