Journal of the Nigerian Society of Physical Sciences

Multimodal transformer networks couple a vision encoder, a learnable projector, and a language decoder to reason jointly across text and image inputs. Dense activation of every feed-forward neuron at every …

With the tremendous growth of cloud computing, cyber threat intelligence systems, and compressed big-data storage, efficient and secure compressed pattern-matching techniques are in high demand. …

Machine learning (ML) models for hepatitis B virus (HBV) outcome prediction must address high variance, bias, and class imbalance. Previous ensemble-learning studies have commonly combined only two models …

Trustworthy machine learning for facies classification depends critically on leakage-free evaluation. This study proposes the Hybrid Explainable Transformer Framework (HETF), combining convolutional neural …

This study classifies 20 rice varieties using three custom-built CNN architectures and four transfer-learning models: MobileNetV2, VGG16, ResNet50, and DenseNet121. The Bangladesh rice dataset, comprising …

The proliferation of agricultural medicines has supported crop productivity, but counterfeit pesticides, herbicides, fertilizers, and bio-based products, fragmented supply-chain management, and limited …

A comprehensive heavy metal pollution assessment of the Taloka and Wurno aquifers within the Sokoto Basin, northwestern Nigeria, was carried out. Field mapping, borehole logging and water quality analysis …

Zero-inflated count data are commonly modelled by mixing a degenerate distribution at zero with a Poisson or negative binomial component, thereby imposing a parametric form on the counts. The transition …

This paper introduces the inverse-order divisor graph associated with a finite group G, a hybrid graph obtained by combining the adjacency conditions of the order-divisor graph and the classical …

The parameterized time neutrosophic soft set supplies a time-indexed family of neutrosophic soft sets together with its basic operations, while the effective-parameter mechanism, which revises an …

This study develops a double Laplace--Sawi decomposition technique for nonlinear partial differential equations by coupling the double Laplace--Sawi transform with Adomian polynomials.  In contrast with a …

This study examines whether Shannon entropy regularization can mitigate the concentration associated with classical minimum-variance portfolio optimization while preserving risk-adjusted performance. Using …

We establish new q-trapezoid-type inequalities for q-differentiable coordinated strongly s-convex functions on bidimensional intervals. A symmetric modulus convention is adopted …

We study how the spectral gap λ₁(G) of a compact metric graph equipped with the Kirchhoff Laplacian depends on its geometric and topological features. First, we examine …

A graph G admits a prime labeling if its vertices can be assigned the integers 1,2,. . . ,|V(G)| so that every pair of adjacent vertices receives coprime labels; a graph with this property is …

In this article, we introduce degenerate multivariate Hermite--based Fubini polynomials by convoluting three-variable degenerate Hermite and Fubini polynomials. First, we derive the generating function …

Rare DNA intermediate structures are essential for biological processes in living organisms, including protein folding and unfolding, gene transcription, and replication. Studies have been conducted on DNA …