This research explores the impact of engagement metrics on predicting customer conversion rates within digital marketing, employing three advanced predictive modeling techniques: Logistic Regression, Decision Trees, and Random Forests. Using a comprehensive dataset of 8,000 customer interactions, the study evaluates critical engagement metrics such as PagesPerVisit, TimeOnSite, and EmailClicks to determine their influence on conversion outcomes. The results indicate that PagesPerVisit and TimeOnSite are the most significant predictors of customer conversion, with the Random Forest model outperforming others, achieving an accuracy of 87.1% and an ROC-AUC score of 0.6979. The Logistic Regression model demonstrated the highest recall for the conversion class at 99.8%, but its performance in predicting non-conversions was less robust, highlighting the challenges of imbalanced datasets. Decision Trees, while offering valuable interpretability, showed a lower accuracy of 79.6% and struggled with precision in identifying non-conversions. These findings suggest that enhancing on-site customer engagement and refining email marketing strategies are pivotal for improving conversion rates. The study contributes to the field of digital marketing analytics by providing empirical evidence on the relative importance of various engagement metrics and offering practical insights for optimizing digital marketing strategies. Additionally, it highlights the benefits of using ensemble methods like Random Forests to achieve more balanced and accurate predictions in customer conversion scenarios.

Advancing times and rapidly developing technology put pressure and responsibility on the management of organizations. Organizational ambidexterity is a concept for an organization that can balance profitability with innovation and development. This study examined the relationship between the triple helix and quality dimensions on organizational ambidexterity mediated by technology readiness and user satisfaction to give management an advantage in addressing this problem. Quantitative analysis methods using PLS-SEM (Partial Least Square-Structural Equation Modeling) were employed in this study. This study was conducted in Indonesia with 425 respondents participating in the data collection, 411 of which were declared valid after filtering. The results of this study demonstrate that the role of the triple helix in developing organizational ambidexterity is very significant and that other variables, such as quality dimensions and technology readiness, also play an essential role. The framework for organizational ambidexterity presented in this study may be helpful for future research in this field. This study can be further developed for future research, especially by adding new external variables that change over time and focusing more on a specific organization. At the very least, this study is relevant for researchers and practitioners to improve business quality using the concept of the triple helix, quality dimensions, and technology readiness.

During Covid-19 pandemic, there was various hoax news about Covid-19. There are truth-clarification platforms for hoax news about Covid-19 such as Jala Hoax and Saber Hoax which categorize into misinformation and disinformation. Classification of supervised learning methods is applied to carry out learning from fact labels. Dataset is taken from Jala Hoax and Saber Hoax as many as 559 data which are made into Class 1 (Misleading Content, Satire/Parody, False Connection), Class 2 (False Context, Imposter Content), Class 3 (Fabricated and Manipulated Content). K-Nearest Neighbor (K-NN) is used to classify categories of misinformation and disinformation. Dissimilarity measure Jaccard Distance is compared with Euclidean, Manhattan, and Minkowski and uses k-value variance in the K-NN to determine the performance comparison results for each test. Results of Jaccard Distance at the value of k = 4 get a higher value than other model with an accuracy 0.696, precision 0.710, recall 0.572, and F1-Score. Maximum results tend to be on label of the most data class in Class 1 (Misleading Content, Satire or Parody, False Connection) with a total of 58 correct data from 61 test data.

Blockchain technology has emerged as a secure and decentralized framework for digital transactions; however, its open and pseudonymous nature also presents significant challenges related to fraudulent activities and malicious nodes. This study investigates the application of machine learning models for blockchain node classification and fraud detection, evaluating three models: Random Forest, XGBoost, and Neural Network. The research leverages a dataset of 10,000 blockchain transactions with 16 attributes, including transaction fees, block scores, stake distribution rates, and coinage. The results demonstrate that machine learning models can effectively classify blockchain nodes with high accuracy. Among the evaluated models, the Neural Network classifier outperformed the others, achieving an accuracy of 95.3%, precision of 95.1%, recall of 95.6%, and an F1-score of 95.3%. Comparatively, XGBoost achieved an accuracy of 94.1%, while Random Forest scored 92.4%. Feature importance analysis highlighted Block Score (0.38), Transaction Fee (ETH) (0.30), and Stake Distribution Rate (0.15) as the most significant factors influencing classification outcomes. Furthermore, confusion matrix analysis revealed that the Neural Network model produced 4780 true positives and 4440 true negatives, with only 200 false positives and 580 false negatives, demonstrating its robustness in identifying fraudulent nodes. Despite these promising results, real-world deployment presents several challenges, including the evolving nature of fraudulent strategies, real-time detection requirements, and scalability concerns. Future research should explore real-time learning techniques, integration of network-based features, decentralized fraud detection mechanisms, and cross-chain anomaly detection to improve model adaptability and effectiveness. By advancing these methods, machine learning-driven fraud detection can contribute to a safer, more transparent, and resilient blockchain ecosystem.

The rapid expansion of metaverse environments has introduced novel opportunities and challenges, particularly concerning user security and trust. This study investigates the application of machine learning techniques to detect scam activities within the metaverse by analyzing user behaviors and interaction patterns. Using a comprehensive dataset, we evaluated three machine learning models—Random Forest, Support Vector Machine (SVM), and Neural Network—for their effectiveness in identifying scams. The Neural Network model achieved the highest performance, with an accuracy of 91%, a recall of 92%, and an AUC of 95%, making it the most reliable solution for this task. Feature importance analysis revealed that attributes such as the number of transactions and average transaction value significantly contribute to scam detection. Hyperparameter optimization further improved model performance, demonstrating the potential of fine-tuned architectures in handling high-dimensional datasets. Despite the Neural Network’s superior performance, its computational complexity highlights the need for lightweight implementations for real-time applications. This research contributes to the growing field of metaverse security by providing a robust framework for scam detection using machine learning. Future work should focus on expanding datasets to capture multi-platform behaviors, incorporating explainable AI (XAI) for improved interpretability, and enhancing model efficiency. These advancements will ensure safer and more trustworthy metaverse ecosystems for users worldwide.

This research employs cluster analysis to elucidate patterns in campus killings across the United States, utilizing a comprehensive dataset spanning over two decades. The study systematically categorizes these incidents into distinct clusters based on geographic, temporal, and demographic criteria to identify underlying patterns and potential risk factors associated with campus violence. Through detailed statistical analysis and visualization techniques, the research reveals significant regional disparities and temporal trends in campus violence, highlighting the concentration of incidents in specific areas and periods. Key findings indicate that campus killings are not uniformly distributed geographically or temporally. Instead, they tend to cluster in certain regions—particularly in the northeastern and central United States—with varying incident frequencies over time. The analysis also uncovers notable demographic patterns, demonstrating that certain racial and socio-economic groups are disproportionately affected. These insights are critical for understanding the dynamics of campus violence and can significantly inform policy-making and preventive measures. The study discusses the implications of these findings for legal frameworks and educational policies, suggesting that more targeted, region-specific interventions could enhance campus safety. By integrating cluster analysis with current legislative and policy contexts, the research provides a foundation for data-driven strategies to mitigate campus violence effectively. However, the study acknowledges limitations related to the data's scope and accuracy, which could impact the generalizability of the findings. Future research directions include expanding the analysis to international contexts, integrating qualitative data, conducting longitudinal studies to assess policy effectiveness, and exploring technological advancements for predictive analytics in campus safety. This research contributes to the academic discourse on campus safety by offering a methodologically robust analysis that links empirical data with policy implications, highlighting the potential for informed legislative actions to foster safer educational environments.