FACULTY OF COMPUTING

This study explores the implementation of Large Language Models (LLMs) for analyzing open-ended survey responses in software engineering. Traditional surveys often focus on structured, multiple-choice questions, which provide quantitative insights but overlook the depth of qualitative developer feedback. To address this limitation, the project designed and implemented an LLM-powered system capable of summarizing responses, detecting sentiments, identifying recurring themes, and revealing open research problems from unstructured text. The system architecture was built using a three-tier design comprising a frontend interface, backend server, and LLM integration layer. A pre-trained model such as GPT was connected via API to process textual data. The study followed a design and implementation-oriented methodology, including data collection from developer surveys, system development, testing, and evaluation. Performance was assessed using both quantitative and qualitative metrics such as accuracy, coherence, and user feedback. Evaluation results demonstrated that the system effectively automated key qualitative analysis tasks with high accuracy and interpretability. However, challenges such as occasional hallucinations, dependency on third-party APIs, and limited dataset scope were noted. Overall, the findings confirm that LLMs can significantly enhance qualitative research in software engineering by providing faster, more consistent, and context-aware insights. The study concludes that integrating LLMs with human oversight presents a promising approach for future software engineering research anddecision-making

Waste management in Edo State suffers from poor communication, irregular service delivery, and weak accountability due to manual record-keeping and fragmented reporting systems. This study developed a web-based Customer Relationship Management (CRM) platform to improve communication, transparency, and citizen participation in waste management. Using the Waterfall methodology, data were collected through interviews and literature review, guiding the design and implementation of the system with ASP.NET Core MVC, C#, Razor Pages, Tailwind CSS, and Microsoft SQL Server. The platform includes a customer interface for registration, request tracking, real-time chat, and announcements, and a waste manager interface for zone management, complaint resolution, and information sharing. Security measures such as Argon2 password hashing and role-based access control were implemented. Testing confirmed full functionality and reliability. The system enhances service delivery bycentralizing communication, standardizing complaint tracking, and promoting citizenengagement, offering a scalable model for technology-drivenwaste management across Edo State.

Color detection is a task that humans perform effortlessly; however, enabling computers to accurately identify colors remains a challenging problem. In many industries, traditional color recognition systems rely heavily on manual processes and paid labor for color-coding items or datasets, which are often time-consuming, repetitive, and proneto human error. To address these limitations, this project presents a deep learning–based color detection program capable of recognizing multiple colors in real time. The system is implemented using Python, a high-level general-purpose programming language, in conjunction with the Open Source Computer Vision Library (OpenCV). By leveraging deep learning techniques, the proposed solution enhances accuracy and efficiency in automated color recognition tasks. The developed system enables computer devices to detect and classify multiple colors in real time, making it suitable for applications across various industries, including pharmaceutical manufacturing, autonomous vehicle development, and robotics. The adoption of this system can significantly reduce production time, minimize reliance on manual labor, and lower operational costs while improving overall productivity

This project presents the design and development of BenTalk, a centralized digital community platform specifically tailored for academic interaction and knowledge sharing within Nigerian universities, with a prototype implementation for the University of Benin's Faculty of Computing. The study addresses the critical challenge of fragmented academic communication, where
students currently rely on general-purpose social media platforms like WhatsApp, Facebook, and Telegram—platforms fundamentally designed for social interaction rather than structured academic discourse. Through comprehensive literature review and system analysis, the research identifies key limitations of existing communication channels: information fragmentation across multiple platforms, lack of institutional oversight, absence of knowledge preservation mechanisms, erosion of professional boundaries, and algorithm misalignment with educational objectives. These challenges necessitate a purpose-built solution that balances the accessibility of social media with the structure required for effective academic collaboration. The BenTalk platform employs a three-tier architecture consisting of a Python FastAPI backend with PostgreSQL database, a React-based responsive web application, and a native Android mobile application developed using Kotlin and Jetpack Compose. The system implements a hierarchical subspace structure organized by departments (Computer Science, Cyber Security, Data Science, Software Engineering, Information and Communication Technology, Information Technology, and Information Science) and academic levels (100L, 200L, 300L, 400L), aligning with the university's existing organizational framework. Core functionalities include secure user authentication via JWT tokens, threaded discussion forums with nested commenting capabilities, upvote/downvote mechanisms for content quality signaling, full-text search across posts, and real-time updates through WebSocket integration. The platform emphasizes knowledge preservation through permanent, searchable archives that
benefit future student cohorts while maintaining intuitive navigation and mobile-first design principles suited to Nigerian infrastructure constraints. The prototype demonstrates technical feasibility and addresses identified gaps in current
academic communication systems. Testing confirms that the platform successfully provides structured academic discourse spaces, reduces information redundancy, facilitates peer-to-peer learning, and enables optional institutional oversight without compromising student ownership of discussions. The system's modular architecture allows for scalability and adaptation to other faculties and institutions. This research contributes to the growing body of literature on educational technology in African
contexts by demonstrating that locally-developed, context-appropriate solutions can effectively address challenges that generic global platforms cannot. The project provides a blueprint for similar implementations across Nigerian universities and offers practical recommendations for institutional adoption, technical enhancement, and sustainable deployment.

The rapid evolution of digital technologies has transformed e-commerce design standards, leaving many early websites outdated and ineffective. This study focuses on the user- centered redesign of a legacy e-commerce interface, using Arngren.net as a case study. The objective was to evaluate the usability and visual experience of the site and to propose a redesign framework that aligns with modern user experience (UX) and interface design (UI) principles. The research adopted a qualitative case study approach, emphasizing heuristic evaluation and comparative analysis. Using Nielsen’s (2020) ten usability heuristics, Arngren.net was assessed for issues relating to layout consistency, navigation flow, accessibility, and visual hierarchy. Findings revealed significant usability flaws, including poor visual organization, low mobile responsiveness, and non-intuitive navigation. These weaknesses were compared with modern e-commerce platforms such as Amazon, eBay, and Shopify-based stores, which prioritize responsive layouts, accessibility compliance, and streamlined user journeys. Based on these insights, a user-centered redesign framework was proposed, integrating simplicity, responsive design, and user trust as key pillars. The redesigned interface emphasizes clear visual hierarchy, improved navigation menus, accessible content structure, and consistency across devices.

The performance quality of web content is a critical determinant of success for e-commerce platforms, directly influencing user engagement and conversion rates. This is especially critical in Nigeria's mobile-first, bandwidth-constrained digital market, where many local websites struggle to meet global performance standards, leading to prolonged load times and a suboptimal user experience that risks customer loss. This study therefore aimed to investigate the performance of selected Nigerian e-commerce websites and diagnose major performance bottlenecks. To achieve this, the project employed a tool-based approach, conducting a performance investigation using Google PageSpeed Insights and GTmetrix to analyze key pages across three major platforms: Jumia, Konga, and Slot. The findings revealed significant performance disparities; Jumia demonstrated relatively optimized performance, Konga exhibited critical inefficiencies with excessively slow load times, and Slot showed inconsistent results, particularly on mobile. In conclusion, the study establishes that technical performance is an inseparable component of web content quality. It provides actionable recommendations, such as image compression and script optimization, underscoring the strategic need for Nigerian e-commerce platforms to prioritize technical excellence to meet up with global standards, enhance user trust, and ensure sustained growth and competitiveness in the evolving digital marketplace.

This work intends to provide a computer based system for the maintenance of personnel records of employees in organization. The work addresses limitations identified with manual method of handling personnel records of employees in a firm by providing a better platform to eliminate fraud, corruption, file hiding and misplacement, records falsification, duplication, fragmentation, inconsistencies and
other vices attendant with manual method of handling personnel records. The Personnel Information System (PIS) software is a user friendly package that gives one the fit to accurately monitor employees' records effortless. With Personnel Information System (PIS), the personnel records of employees in a firm regardless of their place of assignment are simultaneously integrated and rationalized through
the creation of a single system that provides accurate information to all in a time and cost efficient manner. This software is designed for stand-alone windows environment, but has the ability to be networked. The interface is Visual Basic language with the structured query language (SQL). The implementation of the system will provide speedy retrieval of data as well as enhancing effective and efficient data.

This research develops an AI-powered Web Application Firewall (WAF) to detect SQL injection( SQLi) attacks, addressing the limitations of traditional signature-based systems. Using the Kaggle SQLi dataset (30,905 queries), the study applied TF-IDF character-level n-grams and three machine learning models: XGBoost, Random Forest, and SVM, with hyperparameter tuning using grid search and cross-validation. The SVM model performed best, achieving 99.48% accuracy, 99.59% F1-score, 99.90% AUC- ROC, very low false positives and false negatives, and real-time detection with 1.52 ms latency and throughput of 658 queries/second per CPU core. Character n-grams successfully captured common SQLi patterns such as UNION SELECT, OR operators, comments, and tautologies. A Flask-based web application and REST API demonstrated that the system is production-ready, highly scalable, and far cheaper than commercial WAFs. The research confirms that traditional machine learning with good feature engineering can match deep learning performance while remaining simpler and more efficient. Limitations include reliance on one dataset, binary classification, and reduced effectiveness against highly obfuscated or second-order attacks. Future work should involve multi-dataset
testing, adversarial robustness, attack subtype classification, and exploring contextual embeddings. Overall, the study shows that ensemble machine learning provides an accurate, fast, and cost- effective alternative for real-time SQL injection detection.

Small and Medium Enterprises (SMEs) often face challenges in managing the overwhelming number of security alerts generated by their IT systems. Traditional alert systems lack contextual intelligence, leading to alert fatigue, delayed responses, and missed critical incidents. This study presents a context-aware Alert Management System that enhances prioritization accuracy by incorporating operational factors such as alert frequency, entity type, business hours, and historical severity. The system was designed and implemented using a React-based simulation environment with 50 synthetic alerts representing realistic SME security events. Comparative evaluation between a baseline model ((Severity + Criticality)/2) and an enhanced model ((Severity + Criticality + Context Factor)/3) demonstrated a 42.42% reduction in alert fatigue and
complete elimination of false-positive high-priority alerts while maintaining 100% detection of genuine threats. The findings confirm that context-aware alert management significantly improves prioritization accuracy and analyst efficiency. The proposed framework provides SMEs with a cost-effective, transparent, and scalable solution for strengthening their cybersecurity posture and improving real-time incident response.