http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/53 Cyber Security Science Fri, 30 Oct 2026 06:46:16 GMT 2026-10-30T06:46:16Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/32001 Title: Systematic Literature Review on Hate Speech Detection Technology with English and Mixed Language on Social Media Space Authors: Anyaora, Peter; KORMENE, Emily; UDUIMOH, Andrew; OKEAHIALAM, Temple; IKWUAZOM, Callistus; YAKUBU, Lasotte; OFOH, Franklyn; AHMED, Rukayat Abstract: This comprehensive overview of the literature review assesses the developments in hate speech identification with an emphasis on English-language content, the dynamics of hate and offensive speech detection online has evolved. There is need to analyse and indicate these trends to enable researchers to be able to proffer better solutions, however several researchers also consider datasets with mixed languages but no mixed datasets of Nigerian pidgin English or languages. The PRISMA framework is used in the review, which covers papers from 2018 to 2024, to guarantee methodological rigor. Important discoveries emphasize how difficult it is becoming to identify hate speech because of changing linguistic conventions like code-switching and emoji usage. Machine learning techniques, especially deep learning models such as BERT and LSTM, have demonstrated notable achievements in identifying hate speech. However, there are still difficulties in differentiating hate speech from content that is offensive but not hateful because it is more of semantic and contextual rather lexical. In order to increase model generalizability, the paper also emphasizes the necessity of diverse datasets that include various linguistic subtleties and cultural situations. In the end, even though the accuracy of existing models is excellent, there is always a need for better ways to deal with biases and imbalances in the data. In the future, studies focusing on multilingual and mixed-language datasets should investigate interpretability as a means of improving the efficacy and impartiality of hate speech detection algorithms. Thu, 10 Sep 2026 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/32001 2026-09-10T00:00:00Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/31999 Title: Detection of Cyberbullying on Facebook Twitter (X) Using Bi‑ Directional Long Short‑Term Memory and Extreme Gradient Boost Algorithms Authors: Ojeniyi, Joseph; Mohammed, Yusuf; Isah, Abdulkadir; Anyaora, Peter; Olusanjo, Fasola; Uduimoh, Andrew; Baba, Meshach Abstract: The social networking sites have transformed digital communication but have simultaneously enabled the escalation of harmful online behaviors, particularly cyberbullying. This recurring form of digital aggression can lead to serious emotional and psychological harm, including anxiety, depression, and in severe cases, self‑inflicted injury or suicidal behavior. The timely identification and prevention of cyberbullying have become an essential focus of current research. Although numerous machine learning techniques have been applied to detect abusive content, many continue to face challenges such as inefficient kernel tuning, extended training durations, and reduced predictive accuracy. To address these limitations, this study presents a hybrid deep learning architecture that integrates a Bidirectional Long Short‑Term Memory (BiLSTM) network with the Extreme Gradient Boosting (XGBoost) algorithm to improve contextual awareness and classification accuracy. The proposed framework was trained and evaluated on datasets collected from Facebook and X (formerly Twitter), capturing diverse linguistic and behavioral characteristics of user interactions. Experimental results indicate that the BiLSTM–XGBoost hybrid model outperforms conventional classifiers by effectively managing context representation, adaptive learning, and class imbalance. The model achieved 97% accuracy, 95% precision, 92% recall, and an F1‑score of 96%, confirming its robustness and efficiency for cyberbullying detection in dynamic social media environments. The study helps educational institutions, online platforms and legal frameworks provide insights into how to better identify cyberbullying in real‑world scenarios. The study’s high recall ensures that cyberbullies are easily identified and it enhances the understanding of how combining multiple models can lead to better performance in cyberbullying detection. Wed, 01 Jul 2026 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/31999 2026-07-01T00:00:00Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/31998 Title: Ensembled Model of Random Forest and Artificial Neural Network for Short-Term Electric Load Forecasting Authors: Awulu, Benjamin; Ismaila, Idris; Subairu, Sikiru; Anyaora, Peter; Uduimoh, Andrew; Baba, Meshack Abstract: The research focuses on secure short-term electric load forecasting using Random Forest and Artificial Neural Network (RF-ANN) models. Accurate forecasting is crucial for maintaining power system stability and efficiency. Inaccurate forecasting can lead to financial losses, increased load shedding events, and decreased customer satisfaction. Techniques like the State Space Method, Stochastic Time Series, Multiple Linear Regression, General Exponential Smoothing, Knowledge-based Expert Approach, and Machine Learning Approach have been used. The dataset from Transmission Company of Nigeria(TCN) was divided into a 70% training and 30% testing set, and performance accuracies were evaluated using Mean Squared Error, Mean Absolute Error, Root Mean Squared Error, and Mean Absolute Percentage Error. The model achieved an accuracy of 1.73. While RMSE has an accuracy of 1.50899, MSE has 2.27704, and MAE accuracy of 0.00521. The study aims to strike a balance between accurate predictions and ethical data treatment. Fri, 22 Aug 2025 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/31998 2025-08-22T00:00:00Z http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/31997 Title: DEVELOPMENT OF A HYBRID ANOMALY-BASED INTRUSION DETECTION SYSTEM USING AUTOENCODER AND ISOLATION FOREST Authors: Chinedu, Somtochukwu; Uduimoh, Andrew; Anyaora, Peter; Alhassan, John; Yusuf, Hadiza Abstract: As cyberattacks advance in sophistication and fluidity, modern intrusion detection systems (IDS) must progress from static, signature-based models to adaptive models that can detect known and zero-day threats (Labonne, 2020; Ali et al., 2023). This study proposes a hybrid anomaly-based IDS that integrates an Autoencoder (AE) for deep feature representation with an Isolation Forest (IF) for statistical anomaly scoring. The objective is to enhance detection performance for both known and zero-day attacks within an unsupervised learning framework. The proposed model is an adaptation of the hybrid AE–IF framework devised by Mohammed and Telek (2023) and the fog-computing adaptation by Sadaf and Sultana (2020). The hybrid model deploys a logical fusion framework—Hybrid OR and Hybrid AND—dynamically balancing precision and recall in anomaly detection. The model is implemented in Python and trained using unsupervised learning on two benchmark datasets, NSL-KDD and CICIDS2017, so the model operates without any prior knowledge of attack signatures (Engelen et al., 2021). Experimental responses to model anomaly detection capabilities support that the Hybrid OR configuration of the model provided the most balanced anomaly detection performance scores recording F1-score of 0.85 and 0.69 on NSL-KDD and CICIDS2017 datasets respectively—both hybrid performance metrics outperforming the stand-alone AE and IF models. These results are in line with more recent evidence supporting the position that combining feature-reconstruction learning with statistical isolation increases the resilience and adaptability towards network intrusion detection (Elsaid & Binbusayyis, 2024; Alhassan et al., 2024). While performance declined on the more complex CICIDS2017 dataset, the hybrid approach demonstrated improved generalisation relative to individual models. The results suggest that logical fusion of representation learning and isolation-based scoring provides a lightweight and adaptable framework for network intrusion detection, although further validation in live network environments is required before operational deployment (Abuabed et al., 2023; Wang et al., 2023). Mon, 01 Dec 2025 00:00:00 GMT http://irepo.futminna.edu.ng:8080/jspui/handle/123456789/31997 2025-12-01T00:00:00Z