Intrusion Detection Systems (IDSs) play a critical role in protecting modern networks against increasingly sophisticated cyberattacks. However, traditional IDS approaches often suffer from limitations such as low detection accuracy, high false alarm rates, and poor efficiency when processing large-scale network traffic. To address these challenges, this study proposes a hybrid intrusion detection model that combines Random Forest (RF) and Support Vector Machine (SVM) algorithms to improve detection performance and reliability. The motivation for integrating RF and SVM lies in their complementary strengths: Random Forest provides robustness against overfitting and effective handling of high-dimensional data, while SVM offers strong classification capability for complex and non-linear attack patterns. The proposed hybrid RF–SVM model was evaluated using the CIC-IDS2017 dataset, which contains realistic labeled network traffic representing both normal activities and multiple cyberattack categories, including DDoS, brute force, infiltration, botnets, web attacks, and port scanning. Experimental results demonstrate that the proposed hybrid model achieved an accuracy of 94%, precision of 93%, recall of 94%, and F1-score of 94%, outperforming standalone RF and SVM models in intrusion detection performance. Although the false positive rate slightly increased, the proposed approach significantly improved detection sensitivity and reduced the likelihood of missed attacks. The findings indicate that the RF–SVM hybrid model provides an effective and scalable solution for real-time intrusion detection in modern cybersecurity environments.
Modern organizational analytics rely on enterprise data warehouses (EDWs). However, large-scale and centralized AI-driven mining of sensitive data stored in these warehouses means that an organization is vulnerable to privacy leaks, inference attacks, and not complying with regulations. Many of the current privacy-preserving frameworks are not designed for direct integration with OLAP environments. As a result, they cannot provide both high analytical utility and formal privacy protection. The framework proposed in this paper provides a definitive solution through integrating OLAP-based multidimensional feature engineering and federated learning (FL) through FedAvg and differentially private stochastic gradient descent (DP-SGD). The system is designed to partition EDW data across multiple logical clients and allows for collaborative training of a single global neural network while keeping all raw data stored locally within each client. Accuracy of classification results for the UCI Adult Income public benchmark and a synthetic EDW-based dataset tested against four baseline models are reported to be 94.8% with a ε of 2.0 for the synthetic EDW dataset, which is only 1.4% lower than centralized. Furthermore, the use of warehouse-aware federated analytics methods has been shown to reduce membership inference attack (MIA) success rates from 79.4% to 54.8%. Finally, the overall system was maintained with a formal (ε, δ)-differential privacy guarantee, thus validating the solution as being scalable, compliant with regulations, and suitable for the development of privacy-preserving enterprise AI.
Autism spectrum disorder (ASD) is becoming more and more prevalent in the modern era. Using image processing methods, it takes a lot of effort and money to identify autism characteristics through screening testing. The development of machine learning (ML) and artificial intelligence (AI) has made it possible to predict autism at a very young age. Numerous research studies have been conducted using various methodologies, but they have not produced any conclusive findings on the prediction of autism features in relation to age groups. Therefore, the purpose of this study is to design a user interface for predicting ASD in individuals of any age and to offer an effective prediction model based on machine learning techniques. As a result of this study, Random Forest-CART (Classification and Regression Trees) and Random Forest-ID3 (Iterative Dichotomiser 3) and the CNN algorithm were combined to create a prediction model for autism. A user interface was also created based on the suggested prediction model. The AQ-10 (Autism Spectrum Quotient - 10 item version) is a brief screening tool used to identify whether an individual may exhibit traits associated with Autism Spectrum Disorder (ASD). Autism can be diagnosed at any age, although symptoms typically start to show up in the first two years of life and progress over time. Autism sufferers deal with a variety of issues, including learning deficiencies, concentration issues, mental health issues like anxiety and depression, movement difficulties, sensory issues, and many more. ASD-related data, including brain imaging, facial expression recognition, and a dataset of 250 actual datasets gathered from individuals with and without autistic features, were used to assess the suggested model, false positive rate (FPR), sensitivity, and precision. According to the evaluation findings, the suggested prediction model performs better in terms of accuracy and data specificity.
Signature verification and recognition is a critical biometric technique used for identity authentication in various domains, including banking, legal documents, and access control. With the advancements in Artificial Intelligence (AI) and Machine Learning (ML), automated signature recognition systems have become increasingly sophisticated and accurate. This paper provides a comprehensive overview of developing a signature recognition system using Python, AI, and ML techniques. It covers the fundamental concepts and methodologies involved, including data acquisition, preprocessing, feature extraction, model selection, and evaluation. Furthermore, it discusses the role of popular Python libraries and frameworks in implementing such systems and explores common challenges and future directions in the field. The aim is to offer a foundational understanding for studies and practitioners interested in automated signature analysis.
Credit card approval is a critical task for financial institutions that must balance the need for customer acquisition with risk management. Traditional rule-based methods typically lack the flexibility and adaptability of data-driven approaches. This paper presents a machine learning-based framework for predicting credit card approvals using various classification algorithms. Models such as logistic regression, decision trees, random forest, and gradient boosting are evaluated on a publicly available dataset. Performance is measured using accuracy, precision, recall, and F1-score. Results show that machine learning models significantly enhance approval prediction performance and offer valuable insights into feature importance.
