Benin Journal of Physical Sciences, Volume 2 , Issue 2 , Decemeber 2025

The effectiveness of machine learning in classifying prostate cancer from high–dimensional microarray data is often constrained by feature selection instability and scalability challenges. To address these issues, we propose the Ensemble-based Distributed Filtering and Classification Model (EnD-FCM), which integrates stability-driven ensemble learning with distributed feature partitioning.

Stochastic Ordinary Differential Equations (SODEs) are often insoluble via the use of analytic techniques due to randomness that characterize SODEs. Stiffness in SODEs introduces complexity in that numerical method for approximating the solution of stiff SODEs are required to be A-stable, a stringent condition that is attainable by implicit method only. This article presents a class of A-stable methods, that derived via the use of Ito Taylor expansion, Taylor series expansion and method of undetermined coefficients.

An analysis of the impact of the magnetic field, potential depth and the screening parameter on the nature of the energy eigen values of Lead Selenide is done in this study. By using Hulthen-Yukawa Potential Model (HYPM) in an applied magnetic field, the Schrödinger wave equation (SWE) is solved analytically using the Nikiforov-Uvarov  (NU) method to get the solution of the SWE in D-dimensions. With the energy equation obtained, the numerical values were generated with the aid of the Maple software.

Fertilizer application is essential for agricultural productivity but can pose significant risks to soil and groundwater quality when mismanaged. This study employed twodimensional Electrical Resistivity Tomography (2D ERT) to assess fertilizer-related contamination in selected upland and lowland farmlands in Oghara, Delta State, Nigeria. Field data were acquired using the Wenner array configuration of the ABEM Terrameter SAS 300C, with a 10 m electrode spacing and profile lengths of 100–140 m. Data processing and inversion were carried out using RES2DINV software.

This study looked into the chemical composition and antimicrobial properties of essential oil extracted from Aframomum melegueta seeds, commonly known as alligator pepper, using hydro-distillation method. GC-MS analysis showed that the oil contains terpenoids, along with significant phenolic compounds such as guaiacol, catechol, and isoeugenol, compounds known for their antimicrobial and antioxidant properties.

In order to identify distributed denial-of-service (DDoS) attacks on SIP-VoIP  infrastructures in real time, this paper proposes a genetic algorithm-trained modular neural network (MNN) with SMOTETomek balancing. In contrast to traditional firewalls and static filters, our framework leverages modular deep learning and evolutionary optimisation to accurately and adaptively identify malicious traffic.

Two-dimensional Electrical Resistivity Tomography was done on failed portions of Southsouth Benin – Akure Expressway, in Ovia Northeast LGA, to establish the subsurface geological formations responsible for the pavement failure and provide geotechnical support for its maintenance. The survey was done on five worrisome portions of the highway, namely; Igbekhwoe, Odighi, Odiguetue, Omigie and Ugboke communities respectively, using Wenner array electrode configuration to obtain one profiles of 100 m length on both sides of the road at each location.

This research is hinged on the fact that blasting operation, a vital process in the extraction of solid minerals from mines and quarry, with its negative impact on the environment cannot be abandoned. However, with proper ethics and design parameters, the environment can be protected in adherence to the principle of sustainability.

Regions that have numerous geological structures are likely zones that are mineralized. With the aim to map the numerous geological structures dependent on familiar responses shown by various structural units, magnetic data set (aero) of Kabo and its environment was acquired. The magnetic data (aero) was subjected to filtering algorithm with enhancement formula such as the Total or general 2Magnetic Intensity (TMI). Also, we applied Reduction to The Equator (RTE) which assisted in the mapping of numerous geological structures making use of Geosoft Oasis Montaj Software.

A block method of the Adams type is employed to obtain a numerical solution for the boundary layer problems arising from agnetohydrodynamics (MHD) flow. The governing equations are modeled as a set of third and second order differential equations subjected to boundary condition. The block method is implemented to provide efficient numerical approximation of the boundary layer equations. Numerical experiments are carried out on three test problems, and the solutions obtained using the method are compared with existing solutions in the literature.

This study investigates the application of 8% and 10% manganese-doped copper (II) oxide nanoparticles (MDCN) for treatment of contaminated water infested with Salmonella specie, Staphylococcus aureus, lead and cadmium ions. 8% and 10% Mn-doped CuO nanoparticles were synthesized using the co-precipitation method, and their physicochemical properties were characterized using scanning electron microscopy (SEM), X-ray diffraction (XRD), and Fourier Transform Infrared Spectroscopy (FTIR). The antibacterial activity of both nanoparticles was evaluated using a turbidimetric assay.

If an oil reservoir has significant height and is bounded laterally by sealing boundaries, optimizing oil production using a horizontal well requires varying the well location along the open vertical axis, especially in highly anisotropic reservoirs. This work analyzes early, transitional, and late-time dimensionless pressure and its derivative responses for a horizontal well in an anisotropic reservoir that is vertically unbounded but laterally sealed.

When a reservoir is bounded by one or more constant pressure boundaries, such as aquifer or gas cap, the main expectation of the production engineer is to delay the arrival of either water or gas into the well completed in the reservoir to optimize oil production. To achieve this expectation, it is important to understand fluid flow pattern through the reservoir system when oil is produced through the well.

Accurate crop recommendation systems are essential for optimizing agriculturalproductivity and sustainability, yet existing approaches often fail to integrate diverse environmental factors and adapt to location-specific conditions. This study proposes a hybrid machine learning model that leverages soil and climate parameters through a threestage pipeline: Random Forest for feature selection, Extreme Gradient Boosting for robust prediction, and a lightweight Feed forward Neural Network for final decision-making.

Growing interest in the creation of biodegradable composites is a result of the growing demand for sustainable substitutes for synthetic polymers. The main aim of this study is to characterize and contrast biodegradable composites composed of unripe banana starch reinforced with pineapple leaf fiber. Two composite formulations were created: Sample B (10 g of unripe banana starch, 5 mL of glycerol, 100 mL of distilled water, and 5 g of fiber) and Sample A (10 g) of unripe banana starch, 5 mL of glycerol, 100 mL of distilled water, and no added fiber). 

The spreading behavior of automobile oil in soils is a critical factor influencing hydrocarbon contamination, soil degradation, and remediation strategies. This study examined the effects of oil concentration (1500–3000 mg/kg) and time (6–12 hr) on oil spreading rate using a central composite design (CCD) within response surface methodology (RSM). Thirteen experimental runs were conducted, and spreading rates were measured in centimeters. Statistical analysis revealed that the quadratic model provided the best fit for the experimental data (R² = 0.9714; Adj. R² = 0.9510; Pred.