DEPARTMENT OF COMPUTER SCIENCE

Cloud computing has become key to modern digital infrastructure, yet traditional reactive auto-scaling systems struggle to balance performance requirements with cost efficiency. This study addresses the limitations of existing cloud resource management approaches by developing the Predictive and Cost-Optimized Auto-Scaling Framework (PCOAF), a conceptual model that integrates machine learning-based workload forecasting with multi- objective optimization. Through systematic application of Design Science Research Methodology, the research analyzed current auto-scaling systems, identified critical deficiencies including reactive latency, prediction inaccuracy, and cost inefficiency, and designed a three-layered architecture comprising monitoring, prediction and decision, and optimization and execution modules. The framework employs archetype-aware prediction to classify workloads into four behavioral patterns; SPIKE, PERIODIC, RAMP, and STATIONARY enabling tailored scaling strategies for each type. Theoretical validation demonstrates that PCOAF achieves 99.8% workload classification accuracy, reduces mean absolute percentage error to 15%, and projects cost reductions of 22% while decreasing service-level objective violations by 61.4% compared to baseline reactive systems. The study establishes PCOAF's feasibility across five design criteria: relevance, consistency, feasibility, scalability, and economic viability. By addressing identified gaps in both international research and Nigeria's emerging cloud ecosystem, this framework contributes a theoretically grounded and practically applicable solution for intelligent, cost-aware cloud resource management in resource-constrained environments.

The rapid expansion of internet usage and digital commerce has significantly transformed business operations in Nigeria, creating new opportunities for economic growth and entrepreneurship. However, this increased reliance on digital technologies has also exposed online businesses to a wide range of cybersecurity threats, including phishing, malware, ransomware, and denial-of-service attacks. These threats pose serious risks to business continuity, financial stability, customer trust, and national economic development. This study investigates the impact of cybersecurity threats on online businesses in Nigeria, examining the nature, prevalence, and consequences of cyber-attacks on business operations. The study adopts a descriptive research approach and utilizes secondary data obtained from academic journals, industry reports, government publications, and cybersecurity studies. Findings reveal that cyber threats have become increasingly sophisticated, resulting in substantial financial losses, data breaches, reputational damage, and disruptions to online business activities. The study further identifies inadequate cybersecurity awareness, weak security infrastructure, and limited regulatory compliance as major factors contributing to the vulnerability of online businesses in Nigeria. The research concludes that effective cybersecurity strategies, including employee training, implementation of robust security frameworks, regular risk assessments, and stronger regulatory enforcement, are essential for mitigating cyber threats. The study contributes to the growing body of knowledge on cybersecurity in developing economies and provides practical recommendations for business owners, policymakers, and stakeholders seeking to enhance the security and resilience of Nigeria's digital business environment.

As society develops and human motives and relationships change, so does the mission to control
crime and undermine law and order. A perfect society is governed by laws and rules that are all
agreed upon and have quantifiable consequences for any member of the community found to have
floated any specific legal infrastructure unit. Any society's legal framework is complemented by
security tools that carry out the law in an effort to preserve law and order. The legal infrastructure
is the phrase that describes how crucial it is for law enforcement officials to act or defend the
shared interests of people in society. Additionally, it outlines the human rights and civil obligations
of society's members. It is believed that members of the public will notify the appropriate civil and military
defense tools of any incidence involving a breakdown of the constitution and order. In order to arrive at a more
accurate and rational judgment, reported issues are also thought to be carefully collected,
examined, and observed. All of the evidence is gathered, safeguarded, and examined prior to the
accused being charged in court and facing legal action. Without fear or favor, investigations are
carried out. The legislation states that the due prevention procedure used during a criminal
investigation is a crime in and of itself.

Sales forecasting and recommendation systems have become essential tools for businesses seeking to optimize inventory management, enhance customer experience, and maximize revenue. This project focuses on developing a machine learning-based sales forecasting and recommendation system to analyze historical sales data, predict future trends, and provide personalized product recommendations. The forecasting component leverages time series analysis and deep learning techniques such as Long Short-Term Memory (LSTM) networks and ARIMA models to predict future sales with high accuracy. The recommendation system utilizes collaborative and content-based filtering to suggest products tailored to customer preferences. The system is implemented using Python, with data preprocessing, feature engineering, and model training conducted using libraries such as TensorFlow, Scikit-Learn, and Pandas. The recommendation engine is integrated into an interactive user interface that enables businesses to gain insights into customer behavior and optimize their marketing strategies. Through extensive testing and evaluation, the system demonstrates improved forecasting accuracy and enhances the user experience by providing intelligent product recommendations. This project contributes to the field of e-commerce analytics by offering a data-driven solution to boost sales performance and customer engagement

This thesis presents the design and development of an Adaptive Learning Support System that leverages real-time learner analytics, intelligent recommendation techniques, and Explainable Artificial Intelligence (XAI) to enhance personalized education delivery. The proposed system integrates several interconnected modules—including data acquisition, learner state monitoring, adaptive content delivery, personalized recommendations, intelligent interventions, and teacher in-the-loop support—to create a responsive learning environment capable of adjusting to each learner’s unique needs. By analyzing behavioural patterns, content interactions, assessment performance, and contextual factors, the system dynamically recommends suitable learning materials while providing transparent explanations of its decisions. The intelligent services layer ensures scalability, interoperability, and continuous optimization across modules. Overall, the system aims to improve learner performance, engagement, and instructional efficiency, offering a robust and modern approach to adaptive education grounded in computational intelligence, machine learning, and human-centered design principles.

This study looks at how AI-powered chatbots affect student academic performance. It focuses on how these systems impact learning efficiency, engagement, and overall outcomes. As artificial intelligence becomes more common in education, chatbots are increasingly used as virtual learning assistants. They provide instant feedback, personalized study support, and interactive experiences. The research uses both quantitative and qualitative methods to evaluate students’ perceptions, usage patterns, and performance data before and after interacting with chatbots. The findings show that AI chatbots significantly improve students’ understanding of course materials, encourage self-paced learning, and increase access to academic resources. However, the study also points out challenges, such as students becoming too reliant on automated responses and limited development of critical thinking skills. In conclusion, the research emphasizes that while AI-powered chatbots can be effective educational tools, their benefits are greatest when combined with traditional teaching methods and guided human interaction.

This study investigates how transparency and explainability influence user trust in AI-driven interfaces. As AI systems become increasingly embedded in decision-making, users often struggle to understand their processes, leading to skepticism and reduced adoption. Using a mixed-methods approach, data were collected from 62 participants through structured questionnaires assessing transparency, explainability, trust, and user experience. Statistical analyses revealed that higher transparency and clear, human-centered explanations significantly enhance user trust and perceived fairness. However, overly technical or complex disclosures reduce comprehension and engagement. The study proposes a Human-Centered AI framework that integrates adaptive explainability, layered transparency, and user feedback mechanisms. Findings contribute to the growing field of trustworthy AI by offering practical guidelines for designing transparent, ethical, and user-aligned AI interfaces.

The development of a Cloud-based Healthcare Information System (CHIS) is a significant project aimed at creating a scalable and reliable cloud computing system to optimize clinical outcomes and improve patient care. The project covers the entire software development lifecycle, including requirements gathering, system design, implementation, testing, and deployment. Key features of the system include user authentication, patient management, electronic health records (EHR), appointment scheduling, interoperability, telemedicine capabilities, data analytics, and security measures. Technologies such as HTML5, CSS3, JavaScript, React.js, Angular.js, PostgreSQL, MongoDB, Java, Python, and Node.js are used in the system's development. Cloud computing platforms like Google Cloud Platform (GCP), Microsoft Azure, and Amazon Web Services (AWS) are utilized for hosting. Furthermore, the system uses containerization technologies such as Docker and Kubernetes for more effective deployment and management of application components. Standards and procedures from the healthcare sector, including Health Level Seven (HL7) and Fast Healthcare Interoperability Resources (FHIR), are incorporated to ensure flexibility, compatibility, and integrity. Security measures such as access restrictions, audit trails, encryption, and compliance with mandates like the Health Insurance Portability and Accountability Act (HIPAA) are implemented.

This project title is written to help hospitals especially in the areas they encounter problems in keeping their attendance scheme for patient and the solution given to tackle problem such as transforming the existing manual attendance scheme for patients system in which the existing problems involved at the time was laziness of the Doctors to work, misplacement of files, excessive loitering around of patient for their files and loitering of paper in the office. This software reports on our pilot evaluation of AN ELECTRONIC PATIENT MANAGEMENT SYSTEM and their Doctors. The aim is to improve the quality of care to patient and the information about them, as indicated by an improvement in the effectiveness and efficiency of care and in an increase in patient’s satisfaction. This study makes clear that a thorough exploration of users needs before building the system, using qualitative research methods may be crucial because it can prevent data mismatch and maximize the chance that the eventual management system meets its most important aim: to enhance patient’s empowerment and improve the quality of care services. In order to handle this, I decided to introduce a new Electronic Patient Management System for patients. The project dwells more on Computer duty schedule. This is implemented with Visual Basic programming 6.0 language and Microsoft Access for effective information keeping.

A complete system design and implementation abstract framework tailored for your AI-driven spam email detection project is detailed below.AbstractElectronic mail remains a foundational pillar of global digital communication, yet its utility is continuously threatened by the exponential growth of unsolicited messages, phishing attempts, and malware-laden spam. Traditional rule-based and heuristic filters are increasingly obsolete due to their rigidity and inability to adapt to the highly evolving and sophisticated obfuscation tactics employed by modern spammers. To address these critical vulnerabilities, this project designs and implements an intelligent, adaptive Spam Email Detection System leveraging Artificial Intelligence (AI) and Natural Language Processing (NLP) techniques. The primary objective is to build a high-accuracy, real-time pipeline capable of automatically distinguishing legitimate messages ("ham") from malicious content ("spam") with minimal human intervention. The system's structural architecture comprises five core pipelines: data ingestion, text preprocessing, feature engineering, AI model classification, and deployment. During the design phase, raw email datasets (including text body, headers, and metadata) are subjected to rigorous NLP preprocessing, which includes tokenization, stop-word removal, lowercasing, and lemmatization to strip out textual noise. Feature extraction is then executed using TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec embeddings to convert cleaned unstructured text into dense, high-dimensional numerical vectors. For the classification engine, a comparative implementation analysis is conducted using multiple Machine Learning (ML) and Deep Learning (DL) architectures, specifically Naive Bayes, Support Vector Machines (SVM), Random Forest, and Bidirectional Long Short-Term Memory (BiLSTM) networks. The implementation was developed using Python and integrated into a responsive web application dashboard via the Flask framework, allowing users to input raw text or connect live mailboxes for real-time scanning. Experimental evaluation of the models on standard benchmark datasets (such as the Enron and UCI Spam SMS/Email datasets) demonstrates that the BiLSTM deep learning approach, when paired with semantic word embeddings, yields the highest performance, achieving an accuracy rate exceeding 98.2%, exceptional precision, and a drastically reduced false-positive rate. The results prove that integrating AI-driven semantic understanding into email security infrastructure provides a scalable, highly adaptive defensive barrier capable of continuously learning from new spam patterns and significantly hardening enterprise cybersecurity posture.