Article 1 of 2Sayed Mohammad Ashraf Arefi, Mohammad Hamzah Rahimi, Yasin Arad
Abstract
Machine learning (ML) has become a significant interdisciplinary field within artificial intelligence (AI) due to the rapid growth of digital data and the increasing need for data-driven analysis, prediction, and decision support in various domains. The effective application of ML is contingent on problem formulation, data quality, algorithm selection, evaluation, interpretability, computational requirements, security, and deployment conditions. This study aims to provide an analytical overview of ML by integrating four interconnected dimensions: the ML process, major learning algorithms, application domains, and emerging challenges. In particular, it examines the relationship between problem properties, data properties, algorithm selection, model performance, and practical application. An analytical narrative literature review combined with document analysis was adopted. Foundational books, peer-reviewed studies, and selected authoritative technical sources were reviewed and analysed to identify the main learning paradigms, algorithm selection criteria, application patterns, and implementation challenges. According to the review, ML should be viewed as a multi-stage process encompassing problem definition, data acquisition, data preparation, model development, evaluation, deployment, and monitoring. Supervised, unsupervised, and reinforcement learning constitute the principal learning paradigms, while ensemble learning represents a model-combination strategy. Algorithm selection is influenced by task characteristics, data structure and quality, dataset size, computational requirements, evaluation criteria, and interpretability needs. Major challenges include limited or poor-quality data, overfitting, generalization, computational demands, privacy, security, interpretability, and deployment conditions. The study emphasizes that algorithm suitability and ML performance should be assessed within the specific context of the problem, data, and operational requirements.
Keywords: Machine Learning; Artificial Intelligence; Algorithm Selection; Model Interpretability; Data-Driven Decision-Making.
⬇ Download PDF
Article 2 of 2Mohammad Rasoul Popal, Habibullah Andishah, and Abdul Aziz Rahimi
Abstract
Cloud computing has emerged as the key infrastructure for digital transformation, providing scalable and flexible access to computing resources, storage, and software services. Despite its major advantages such as scalability, flexibility, lower infrastructure costs, and on-demand access to computing resources, widespread adoption of cloud computing has introduced significant security concerns regarding the confidentiality, integrity, and availability of information. As a result, the development of intelligent approaches to assigning security risk has become an important requirement for cloud computing environments. The objective of this study is to develop an intelligent model for the assessment of security risk in cloud computing environments, using supervised machine learning algorithms. First, a systematic literature review was conducted to identify and classify the major factors affecting safety risk. The identified factors were evaluated and prepared for model development, resulting in a set of 18 safety risk factors. Three supervised machine learning algorithms—Decision Tree, K-Star (K*), and Random Filtered Classifier—were then evaluated and compared to predict the level of safety risk. To identify relevant variables and to support the model development, three methods for selecting features, namely Best First Search, Ranker, and Random Search, were applied using the WEKA and Orange Data Mining environments. The performance of the model was evaluated using the Root Mean Square Error (RMSE) and correlation coefficient (R) under different training–testing data-split scenarios. The results showed that the 95% training and 5% testing split produced the lowest RMSE among the evaluated data-split scenarios. The Decision Tree algorithm achieved the lowest RMSE across the evaluated scenarios, reaching 0.0018 under the 95% training and 5% testing split. The Random Filtered Classifier and K-Star algorithms produced identical RMSE values across the evaluated scenarios. These findings demonstrate the applicability of supervised machine learning algorithms to intelligent security risk assessment and provide a basis for developing data-driven approaches to security risk analysis in cloud computing environments.
Keywords: Cloud Computing; Cloud Security; Security Risk Assessment; Supervised Machine Learning; Feature Selection; Decision Tree; K-Star; Risk Prediction.
⬇ Download PDF