1 March 2023 in Indonesian Journal of Electrical Engineering and Computer Science
Competitive asset markets and increased globalization have continued to ripple the food value chain with complex dynamics, which has led to a range of challenges such as food safety and quality, traceability, and overall supply chain inefficiency. These have further continued to endanger the general well-being of society. With rice as a staple food in Nigeria, the rice food supply value chain consists of a series of tasks, processes, and activities that are linked together from freshly harvested products to consumer demand and supply. Study advances the SmartRice, a sensor-based block-chain framework that decentralizes as well as provides a decision-support for the food supply value chain process by first ensuring that accurate data of harvested goods are reported, and passed on to a chain. The study advances a decentralized framework to eliminate various forms of fraud rippled across the existing centralized system, minimize corruption through its sensor-based layered model as well as minimize the error in reported data along the value chain.
1 December 2022 in Indonesian Journal of Electrical Engineering and Computer Science
In our exploratory quasi-experimental study, 480-student were recruited and exposed to social engineering directives during a university orientation week. The directives phishing attacks were performed for 10 months in 2021. The contents attempted to elicit personal user-data from participants, enticing them to click compromised links. The study aimed to determine cybercrime risks among undergraduates in selected universities in Nigeria, observe responses to socially-engineered attacks, and explore their attitudes to cybercrime risks before/after such attacks. The study generalized that all participants have great deal awareness of cybercrime, and also primed all throughout study to remain vigilant to scams. The study explores various types of scam and its influence on students’ gender and age on perceived safety on susceptibility to phishing scams. Results show that contrary to public beliefs, none of these factors were associated with scam susceptibility and vulnerability rates of the participants.
1 April 2023 in International Journal of Electrical and Computer Engineering (IJECE)
Access ease, mobility, portability, and improved speed have continued to ease the adoption of computing devices; while, consequently proliferating phishing attacks. These, in turn, have created mixed feelings in increased adoption and nosedived users’ trust level of devices. The study recruited 480-students, who were exposed to socially-engineered attack directives. Attacks were designed to retrieve personal data and entice participants to access compromised links. We sought to determine the risks of cybercrimes among the undergraduates in selected Nigerian universities, observe students’ responses and explore their attitudes before/after each attack. Participants were primed to remain vigilant to all forms of scams as we sought to investigate attacks’ influence on gender, students’ status, and age to perceived safety on susceptibility to phishing. Results show that contrary to public beliefs, age, status, and gender were not among the factors associated with scam susceptibility and vulnerability rates of the participants. However, the study reports decreased user trust levels in the adoption of these new, mobile computing devices.
1 October 2024 in Bulletin of Electrical Engineering and Informatics
With the incessant outbreak of fire, the heavy loss to both lives and properties in the society fire has since become a critical issue and challenge that needs our daily attention to be resolved. Loss of lives and properties to fire outbreak in 2021 alone as occurring in major Nigerian markets and residential homes was estimated at over 3 trillion Naira. Our study proposes a wireless sensor network internet of things (IoT) based ensemble to aid the effective monitoring, detection and alerting of residents and fire service departments. With cost as a major issue and the requisite installation of fire and smoke detectors in many houses our ensemble can efficiently integrate into the existing system using the ESP8285-controller to create a comprehensive access control system. The system provides real time monitor and control capabilities that will allow administrators to track and manage fire monitor and detection within a facility. Thus, enhances system's efficiency and performance.
1 March 2024 in European Journal of Medical and Health Research
Counterfeit drugs are a problem both in developed and developing countries. Nigeria is a developing country experiencing the problem of counterfeit drugs and as such is the use case for this research work. The negative effects counterfeit drugs have on human health and economy of the country calls for urgent attention. One amongst many reasons why counterfeit drugs are still in circulation in Nigeria is the unmonitored drug distribution system. Drugs throughout their shelf life moves from one owner to another, from the importer/manufacturer to Distributor, wholesaler to government/private healthcare sectors and then to the patients. The current drug distribution system in Nigeria is a manually, unmonitored system, where the regulatory agency body National Agency for Food and Drug Administration control (NAFDAC) does not have detailed record on the movement of drugs from either the importer or manufacturer to the patient in the system. In this study blockchain technology framework that can handle the drug distribution chain was discussed, since blockchain technology offers immutability, transparency and tamper proof, it functionalities are what intrigued the study as it can help monitor the movement of drug in the distribution chain. The proposed system is a prototype demonstrating the hashing mechanism with a deployed smart contract and how it can help in showing the authenticity of a drug. The blockchain framework is a private blockchain, where participants on the network are permitted to have access to the network by the regulatory agency. The system was evaluated based on it’s effectiveness from users reports using the security, user friendly, response time, flexibility, data integrity checks, robust, hashing mechanism functional as the benchmark. The actual test result equates the expected test result, showing that the system is efficient.
8 May 2025 in Asian Journal of Research in Computer Science
High-dimensional datasets pose significant challenges in machine learning, including overfitting, increased computational complexity, and reduced model interpretability. In response to these challenges, this paper introduces the Three Conditions for Feature Aggregation (3ConFA) framework, a feature selection approach based on ensembles that seeks to reduce dimensionality without reducing predictive performance. The 3ConFA framework integrates three key techniques; Chi-square (χ²) test, Information Gain (IG), and Decision Tree-based Recursive Feature Elimination (DT-RFE), and applies stringent conditions to select only the most relevant features. A feature is retained if and only if it satisfies all three conditions: (1) IG score ≥ mean IG threshold, (2) χ² score ≥ mean χ² threshold, and (3) DT-RFE importance score = 1. Experimental evaluation on ten benchmark datasets demonstrates the effectiveness of the framework, achieving feature reductions of up to 98.75% (e.g., Dexter dataset) while improving classification accuracy (e.g., Madelon: 78% → 85%). Performance metrics (precision, recall, F1-score) were consistently enhanced after feature selection, confirming that 3ConFA enhances model generalization without tradeoff in critical information. The framework’s adaptability across diverse datasets highlights its potential for applications in fraud detection, healthcare, and image classification. This study contributes a structured, condition-driven feature aggregation approach that outperforms traditional filter and wrapper methods. Future work may explore adaptive thresholding and integration with deep learning models.