ADVANCED JOURNAL OF ENGINEERING AND SCIENTIFIC APPLICATIONS

MACHINE LEARNING APPROACHES TO FRAUD DETECTION AND RISK ANALYTICS IN GSM-BASED FINTECH SYSTEMS IN NIGERIA

Joseph Osahon Idemudia, Taiwo Adisa Oyeniran — Tue, 13 Jan 2026 00:00:00 +0000

The increasing adoption of mobile phone numbers as primary identifiers in financial transactions has significantly transformed the operations of financial technology (FinTech) companies in Nigeria. Leveraging Global System for Mobile Communications (GSM) numbers as account identifiers has enhanced access to digital financial services while introducing challenges related to security, data management, and analytics. This study investigates the application of data mining and machine learning techniques in GSM-based FinTech systems, with emphasis on fraud detection, anomaly detection, customer segmentation, and credit risk assessment. A synthetic yet statistically representative GSM-linked transaction dataset was developed to reflect real-world FinTech operations under regulatory and ethical constraints. Controlled modeling experiments were conducted using supervised and unsupervised learning techniques. Experimental results show that the Random Forest classifier achieved an accuracy of 96.8%, precision of 94.5%, recall of 92.7%, and an F1-score of 0.935, substantially outperforming traditional rule-based approaches. Receiver Operating Characteristic analysis yielded a high area under the curve exceeding 0.94, indicating strong discriminative performance. Anomaly detection models effectively identified SIM-swap and account takeover fraud, achieving detection rates above 90% with average detection delays below two minutes. Unsupervised clustering revealed four distinct customer segments, including a high-risk group representing approximately 12% of users. In addition, predictive credit risk models achieved 92.1% accuracy and reduced simulated loan approval time by 37%. Feature importance analysis identified GSM-specific behavioral indicators—such as transaction value, SIM change events, and location instability—as key predictors of fraudulent and high-risk behavior. Overall, the results demonstrate that data mining significantly enhances security, operational efficiency, and financial inclusion in GSM-based FinTech systems.

IMPACT OF BOKO HARAM INSURGENCY ON CONSTRUCTION PROJECT DELIVERY IN BORNO STATE, NIGERIA

Sat, 14 Mar 2026 00:00:00 +0000

Construction project delivery in conflict-affected regions is often constrained by security challenges. This study examines the impact of the Boko Haram insurgency on construction projects in Borno State, Nigeria, focusing on site accessibility, workforce safety, material supply, and institutional support. A descriptive survey design was adopted, with primary data collected through structured questionnaires and semi-structured interviews of 70 purposively sampled stakeholders, including engineers, architects, project managers, contractors, consultants, and government officials drawn from a population of 233 using Yamane’s formula. Data were analyzed using descriptive statistics and Likert scale analysis, presented as percentages, tables, and graphs. Findings indicate that insecurity significantly disrupts project delivery: 73.8% of respondents strongly agreed that access to sites is reduced, 76.9% that workers are reluctant to report due to security threats, 80.0% that insurgent activities slow progress, and 70.8% that military escorts are often required. Additionally, 64.6% strongly agreed that insecurity contributes to the use of inferior materials, affecting project quality. The study concludes that Boko Haram insurgency imposes substantial operational and logistical constraints on construction stakeholders in Borno State. Enhancing security coordination, improving institutional support, and adopting conflict-sensitive project management strategies are essential for improving project delivery in insurgency-affected regions.

EVALUATION OF INDEX PROPERTIES OF BLACK COTTON SOIL STABILIZED WITH SOYA BEAN HUSK ASH

E.O Ogundele, J.E. Sani, G. Moses, K.K. Mustapha, O.K. Kevin — Sun, 15 Mar 2026 00:00:00 +0000

The objective of this study was to assess the index characteristics of black cotton soil stabilized with Soybean Husk Ash (SBHA). The black cotton soil used in the study was obtained from the Gamadadi region of Borno State, Nigeria. The Atterberg limit test was conducted to determine the liquid limit, plastic limit, plasticity index, and linear shrinkage. In addition, sieve analysis was carried out. The soil was classified as CH (clay of high plasticity) under the Unified Soil Classification System (USCS, ASTM 1992) and as belonging to the A-7-6(0) group under the AASHTO (1996) classification system. When conditions change from wet to dry, the soil color appears grayish-black. The index properties of the soil showed that the liquid limit was 52%, which decreased to 46% at 16% SBHA. Similarly, the plastic limit decreased from 19.52% to 16.37% at 16% SBHA. This occurred because higher-valence cations replaced weakly bonded ions in the clay, resulting in particle flocculation and the release of absorbed water. As a result of pozzolanic reactions and clay dilution, the plasticity index decreased from 32.48% to 29.63%, indicating reduced clay activity and improved soil stability. At 16% SBHA, the specific gravity decreased from 2.63 to 2.38 and the shrinkage limit decreased from 22.86% to 14.29%. While the Optimum Moisture Content increased from 18.41% to 24.54%, the Maximum Dry Density decreased from 1.68 g/cm³ at 0% SBHA to 1.58 g/cm³ at 12% SBHA and then slightly increased to 1.59 g/cm³ at 16% SBHA. To improve the index properties of black cotton soil, an optimum SBHA content of 12% is recommended.

A SURVEY OF RENEWABLE ENERGY INTEGRATION IN OIL AND GAS SYSTEMS: DEVELOPMENTS, CHALLENGES AND FUTURE DIRECTIONS

S. M. Lawan, H. Lawan, Idris Saadu Idris, A.Y. Muhammad, I.D. Umar — Thu, 26 Mar 2026 00:00:00 +0000

The integration of renewable energy into the operations of the oil and gas sector has emerged as a critical trend driven by increasing environmental concerns, stringent regulatory requirements, and rapid technological advancements. This paper presents a comprehensive review of the current trends, key challenges, and future directions associated with renewable energy adoption within the oil and gas industry. The analysis reveals a significant transition toward the incorporation of renewable energy sources, including solar, wind, and hydrogen, into conventional oil and gas operations. This shift is primarily motivated by the need to reduce greenhouse gas emissions, enhance operational efficiency, and diversify energy portfolios in line with global sustainability objectives. Furthermore, the study highlights that renewable energy integration offers multiple benefits, such as improved energy security, optimized resource utilization, and potential cost savings through reduced dependence on fossil fuels and access to supportive policy incentives. Despite these advantages, several challenges persist, including the intermittency of renewable sources, grid integration complexities, and infrastructure limitations, which may hinder large-scale deployment. Looking forward, advancements in renewable energy technologies, coupled with declining costs and supportive regulatory frameworks, are expected to accelerate adoption across the sector. However, the pace of integration will largely depend on the ability of industry stakeholders to address technical, economic, and policy-related challenges. Overall, renewable energy integration represents a transformative pathway for the oil and gas sector, enabling companies to mitigate environmental risks while enhancing long-term competitiveness in an evolving global energy landscape. This survey adopts a systematic and analytical review methodology, Data Collection: Peer-reviewed journals (IEEE, Elsevier, Springer, MDPI), and Industry reports.

IMPROVEMENT OF CRUDE OIL–CONTAMINATED LATERITIC SOIL USING STEAM INJECTION AND PERIWINKLE SHELL ASH FOR ROAD SUBBASE APPLICATIONS

J.E. Sani, Z.I. Ummisawa, G. Moses, O.K. Kevin — Wed, 08 Apr 2026 00:00:00 +0000

Contamination from petroleum hydrocarbons poses a significant threat to soil and groundwater systems, thereby endangering ecosystems, human health, and socio-economic sustainability. This study evaluates the combined effectiveness of steam injection remediation and periwinkle shell ash (PSA) stabilization in improving the geotechnical properties of crude oil–contaminated lateritic soil for road subbase applications. Steam injection was applied at durations of 1 hour, 1 hour 30 minutes, 2 hours, and 2 hours 30 minutes. The remediated soil was subsequently stabilized with PSA at incremental additions of 4% up to 16%. Geotechnical tests were conducted in accordance with BS 1377. Results showed that steam injection achieved a maximum total petroleum hydrocarbon (TPH) removal efficiency of 75.89% at 1 hour 30 minutes, beyond which efficiency declined. Compaction results indicated that the untreated soil had Maximum Dry Density (MDD) values of 1.60 Mg/m³ for British Standard Light (BSL) and 1.75 Mg/m³ for British Standard Heavy (BSH), which increased to peak values of 1.69 Mg/m³ and 1.88 Mg/m³, respectively, after treatment. Optimum moisture content (OMC) increased with PSA content, reaching 16.56% for BSL and 14.65% for BSH at 12% PSA. The untreated soil recorded an unsoaked CBR of 19.3%, while peak values of 65.78% for BSL and 83.48% for BSH were obtained at 8% PSA with 1 hour 30 minutes of steam injection. Soaked CBR improved from 6.81% to 31.82% for BSL and 49.64% for BSH, satisfying Nigerian General Specifications for subbase materials. The study demonstrates that the combined use of steam injection and PSA provides a sustainable and effective method for improving crude oil contaminated lateritic soils for road subbase material.

COMPARATIVE ANALYSIS OF TRADITIONAL RFID TAG ANTI-COLLISION TECHNIQUES FOR IOT APPLICATIONS

Umelo N. H., Eze U. J., Eleje N.E. — Wed, 08 Apr 2026 00:00:00 +0000

Radio Frequency Identification (RFID) is seen as inevitable especially within the sensing layer of IoT architecture. RFID is important because it communicates wirelessly and numerous tags can be identified at same time. Besides, low cost of passive RFID tags makes the tagging and connectivity of things around us realizable. The tag collision problem (TCP) of RFID systems have received much research attention for small-scale tag deployments. However, the overcrowded nature of RFID tags in IoT puts a strong doubt, the ability of existing RFID tag anti-collision methods to effectively tackle TCP. This paper examines traditional methods (the probabilistic and deterministic methods) used in developing the existing RFID tag anti-collision protocol. Their pros and cons were explored. Both methods were implemented as M-files in MATLAB while simulation analysis were performed with varying tag density. RFID tags’ unique RN16 were simulated using 16-bit random numbers generated in MATLAB and under overcrowded environment of minimum 100 tags and maximum 1000 tags. The 1000 tags represents IoT scenario. Results of a Monte-Carlo simulation show a tag identification rate (TIR) of 0.98 in IoT for ALOHA-based (probabilistic) method and a tag identification rate of 0.90 in IoT for Tree-based (deterministic) method. Simulation result also shows ALOHA-based method uses less number of time slots (identification time) of 3200 than 4400 of Tree-based method. Since access to the communication channel is represented in times slots, the simulation results imply that ALOHA method takes less time to identify tags.