Financial fraud negatively impacts organizational administrative processes, particularly affecting owners and/or investors seeking to maximize their profits. Addressing this issue, this study presents a literature review on financial fraud detection through machine learning techniques. The PRISMA and Kitchenham methods were applied, and 104 articles published between 2012 and 2023 were examined. These articles were selected based on predefined inclusion and exclusion criteria and were obtained from databases such as Scopus, IEEE Xplore, Taylor & Francis, SAGE, and ScienceDirect. These selected articles, along with the contributions of authors, sources, countries, trends, and datasets used in the experiments, were used to detect financial fraud and its existing types. Machine learning models and metrics were used to assess performance. The analysis indicated a trend toward using real datasets. Notably, credit card fraud detection models are the most widely used for detecting credit card loan fraud. The information obtained by different authors was acquired from the stock exchanges of China, Canada, the United States, Taiwan, and Tehran, among other countries. Furthermore, the usage of synthetic data has been low (less than 7% of the employed datasets). Among the leading contributors to the studies, China, India, Saudi Arabia, and Canada remain prominent, whereas Latin American countries have few related publications.

One of the fields where Artificial Intelligence (AI) must continue to innovate is computer security. The integration of Wireless Sensor Networks (WSN) with the Internet of Things (IoT) creates ecosystems of attractive surfaces for security intrusions, being vulnerable to multiple and simultaneous attacks. This research evaluates the performance of supervised ML techniques for detecting intrusions based on network traffic captures. This work presents a new balanced dataset (IDSAI) with intrusions generated in attack environments in a real scenario. This new dataset has been provided in order to contrast model generalization from different datasets. The results show that for the detection of intruders, the best supervised algorithms are XGBoost, Gradient Boosting, Decision Tree, Random Forest, and Extra Trees, which can generate predictions when trained and predicted with ten specific intrusions (such as ARP spoofing, ICMP echo request Flood, TCP Null, and others), both of binary form (intrusion and non-intrusion) with up to 94% of accuracy, as multiclass form (ten different intrusions and non-intrusion) with up to 92% of accuracy. In contrast, up to 90% of accuracy is achieved for prediction on the Bot-IoT dataset using models trained with the IDSAI dataset.

In this study, the performance of different unsupervised machine learning algorithms used for intrusion detection within unbalanced data environments were analyzed; these algorithms included the K-means++ algorithm, density-based spatial clustering of applications with noise (DBSCAN), local outlier factor (LOF), and isolation forest (I-forest) using the BoT–IoT dataset. Performance metrics such as purity, homogeneity_score, completeness_score, v_measure_score, and adjusted_mutual_info_score were used to evaluate the effectiveness of algorithms in detecting various types of attacks such as distributed denial of service (DDoS), denial of service (DoS), and reconnaissance. Similarly, different methods were used for the automatic selection of the optimal number of clusters such as the elbow method, silhouette coefficient, Calinski–Harabasz index, and Davies–Bouldin index. Moreover, principal component analysis (PCA) was used to explain data variance and the influence of variables on intrusion detection. Results revealed that the K-means algorithm achieved 95% purity as well as 95% and 99% prediction accuracies for normal and abnormal data, respectively. The I-forest algorithm achieved 95% purity as well as 99% and 90% prediction accuracies for normal and abnormal data in a balanced dataset, respectively. These findings indicated that I-forest exhibited a low central processing unit (CPU) consumption rate of 10% on balanced data, outperforming DBSCAN, K-Means++, and LOF, with 16% consumption rates.

In Wireless Sensor Networks (WSN), nodes are vulnerable to security attacks because they are installed in a harsh environment with limited power and memory, low processing power, and medium broadcast transmission.Therefore, identifying threats, challenges, and solutions of security and privacy is a talkingtopic today. This article analyzes the research work that has been carried out on the security mechanisms for the protection of WSN against threats and attacks, as well as the trends that emerge in other countries combined with future research lines. From the methodological point of view, this analysis is shown through the visualization and study of works indexed in databases such as IEEE, ACM, Scopus,and Springer, with a range of 7 years as an observation window, from 2013 to 2019. A total of 4,728 publications were obtained, with a high rate of collaborationbetween China and India. The research raised developments, such as advances in security principles and defense mechanisms, which have led to the design of countermeasures in intrusion detection. Finally, the results show the interest of the scientific andbusiness community in the use of artificial intelligence and machine learning (ML) to optimize performance measurements.