http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/104 Thu, 30 Apr 2026 05:44:07 GMT 2026-04-30T05:44:07Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30080 Title: A Machine Learning Approach to Fake News Detection Using Support Vector Machine (SVM) and Unsupervised Learning Model Authors: Hosea, I. G.; Waziri, V. O.; Ismaila, Idris; Ojeniyi, Joseph Adebayo; Olalere, Morufu; Adebayo, O. S. Abstract: Blogging over the years have become a lucrative business, the bloggers main aim is to attract people to his or her blog. In the quest for that, many blogs or page post fake news by using enticing captions to captivate the minds of readers. The captions are mostly displayed on social media and by clicking on the captions, the reader will be redirected to the blog where the news is been posted. The posted fake news can sometimes lead to misinformation to the public, violence, inciting conflict and extreme cases, death. Many works have been done on fake news detection with good accuracy rate in terms of detecting fake news. This paper presents an effective way of detecting fake news using Support Vector Machine (SVM) and Lagrangian Duality which yielded an accuracy of 95.74%. Sun, 01 Jan 2023 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30080 2023-01-01T00:00:00Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30079 Title: Learners’ Privacy-Preserving Scheme for Ranking Data Sensitivity in Mobile Learning System Authors: Muhammad, M. K.; Olaniyi, O. M.; Osang, F.; Oyefolahan, I. O.; Ojeniyi, Joseph Adebayo; Kolo, I. M. Abstract: Mobile Learning System (MLS) is facing new challenges in terms of privacy, such as data collection, storage, and sharing because of the core infrastructure and network that enables cloud computing services. Most of the data exchange in Mobile Learning System (MLS) require mandatory authorisation to allow access to the learners’ information in the MLS. Therefore, this article attempts to rank learners' sensitive attributes stored in MLS. Thus, concerns about privacy breaches motivated this paper to adopt an attributes partitioning strategy into the sensitive attributes to enforce privacy during learners’ profile information access. The article adopted the informed consent phenomenon to determine and formulate learners’ data privacy attributes sensitivity using the Fuzzy Analytic Hierarchy Process (FAHP) Algorithm. Results from the implemented Learners’ Privacy Preserving (LPP) Algorithm determined normalized weights of top-five rank-selected learners’ sensitive data to include: Browsing History (1ST, Ranked), Geolocation Data (2ND, Ranked), IP Address (3RD, Ranked), web Browser (4TH, Ranked), Medical Records (5TH, Ranked) and CGPA (10TH, Ranked) respectively. This indicates that these five most sensitive features are at risk and require protection to prevent privacy breaches, thus ensuring privacy preservation that prevents unauthorised access to learners’ sensitive data in the mobile learning system environment. The ranking of sensitive data in this paper could serve as inspiration for future research work on mobile learning security to improve the privacy of sensitive attributes in MLS environment Mon, 01 Jan 2024 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30079 2024-01-01T00:00:00Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30078 Title: Stack Ensemble Model For Detection Of Phishing Website Authors: Wabi, H. A.; Ojeniyi, Joseph Adebayo; Ismaila, Idris; Subairu, S. O. Abstract: The importance of mitigating the risk associated with online purchases for retail businesses cannot be overstated, phishing websites are a significant threat to these businesses as they deceive web users and steal their sensitive information. In this paper, we proposed a stack ensemble model consisting of three different classification algorithms RandomForest, GradientBoosting(Xgboost), and Adaptive Boosting (Ada Boost), with logistic regression as an aggregator, to detect phishing websites. The proposed model was evaluated and compared with each individual algorithm in the stack, as well as with existing studies. Our analysis showed that the proposed stack ensemble model performed better than each algorithms and existing studies in terms of accuracy, recall, specificity, precision, and F1 score. Specifically, our proposed model obtained an accuracy of 98.72%, recall of 98.84%, specificity of 98.60%, precision of 98.60%, and F1 score of 98.72%, with a low error rate of 1.28%. Overall, the results of our study demonstrate the effectiveness of the proposed stack ensemble model in detecting phishing websites for retail businesses. This model can play a crucial role in enhancing cybersecurity measures and mitigating the risks associated with online purchases. The findings of this study could be beneficial to security experts and policymakers working to improve cybersecurity in the retail industry. Mon, 01 Jan 2024 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30078 2024-01-01T00:00:00Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30077 Title: Comparative analysis of SVM kernels for DDOS attack detection in SDN Authors: Abdullahi, Aishatu Wabi; Ismaila, Idris; Olaniyi, O. M.; Ojeniyi, Joseph Adebayo; Adebayo, O. S.; Mishra, A Abstract: Software Defined Networks (SDN) were created to address the scalability and high maintenance costs of conventional networks. SDN is susceptible to Distributed Denial of Service (DDOS) attacks because of its scalability, programmability, and centralized control. DDOS aims to deplete the target host's resources and block authorized users from accessing it. DDOS attack detection in SDN has been accomplished using Support Vector Machine. A supervised machine learning technique called SVM searches for the hyperplane that best divides two classes. SVM features a number of kernel functions and regularization parameters that, if not properly chosen, can have an impact on performance. This study looks at some of the regularization and SVM Kernel functions that are already available, and how they function with various parameter values. The outcome demonstrates that various criteria produced diverse outcomes. The performance of the polynomial and Gaussian kernels was superior, but their computational cost was higher Mon, 01 Jan 2024 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/30077 2024-01-01T00:00:00Z