Journal of Mathematical Sciences and Informatics

DYNAMICAL BEHAVIOUR OF A DISCRETE-TIME PREDATOR-PREY MODEL WITH INTRAGUILD PREDATION

2025-04-05T15:31:02+08:00

This article examines a discrete-time predator-prey model that incorporates intraguild predation. For biological reasons, positivity and boundedness of solutions are verified. A condition for the stability of an interior fixed point is derived. Global stability criterion of the interior fixed point is obtained. It is found that the system exhibits Neimark-Sacker and period-doubling bifurcations under certain restrictions on the system parameters. The system shows a chaotic nature for a particular choice of system parameters. This phenomenon may be prevented by applying a hybrid control technique. We have observed that an increase in the amount of intrinsic growth rate of prey initially destabilises the system and finally stabilises, whereas an increase in the amount of intraspecific competition rate of prey cannot stabilise the system. Some illustrations support our analytical findings.

SEA TURTLE-PREDATOR MODEL WITH THE EFFECT OF EXPLOITATION PARAMETER

2025-04-28T11:12:45+08:00

Sea turtles, vital to marine ecosystems, face considerable threats due to human exploitation, putting their global populations at risk. This study examines a prey-predator model to explore the dynamics of sea turtle populations, focusing on the impact of humans and their interactions with marine predators. The goal is to examine how the dynamics of this model change when the exploitation parameter is altered. Through stability and bifurcation analyses, we assess the conditions that determine whether sea turtle populations remain stable or undergo substantial changes as exploitation pressure fluctuates. We employed eigenvalue analysis to establish the model’s stability conditions. Furthermore, bifurcation analysis reveals the system’s complex behaviour as the exploitation rate changes. By constructing bifurcation diagrams, we pinpoint critical values at which significant shifts in the dynamics of sea turtle populations occur, including the emergence of alternative stable states or the onset of population decline. The results revealed that once the exploitation rate is below its critical threshold, both the sea turtle as well as predator populations survive and oscillate. Nonetheless, if exploitation is too high, both populations face extinction within 20 to 60 years. Therefore, this research is important for raising awareness about the value of sea turtles and the threats they face, which can help reduce exploitation.

TOPOLOGICAL INDICES OF LINE AND PARALINE GRAPHS OF CONDUCTIVE 2D METAL-ORGANIC FRAMEWORKS (MOFs)

2025-10-05T23:03:39+08:00

Topological indices assign a numerical value to a chemical structure. The use of these graph indices in chemical graph theory is very broad. In this article, we calculate several well-known degree-based topological indices for the chemical structures of conductive 2D metal-organic frameworks (MOFs) by applying the concept of line and paraline graphs. These results are instrumental in the design of emerging networks, enabling the study of their topological indices to gain a deeper understanding of their underlying topology.

THIN FILM BOUNDARY LAYER FLOW OF WATER-BASED HYBRID NANOFLUIDS ALONG WITH AN APPLIED MAGNETIC FIELD: AN ANALYTICAL APPROACH

2025-10-05T18:13:41+08:00

This research presents an analytical examination of the steady-state boundary layer flow and thermal characteristics of hybrid nanofluid
systems. Both nanofluid and hybrid formulations are analysed; a similarity transformation recasts governing partial differential equations
into ordinary differential forms. The derived system is tackled using the Homotopy Asymptotic Method (HAM), which provides closed-form analytical solutions that contribute to an understanding of the flow behaviour. The researcher produces visual representations of the flow and temperature fields for the different cases and performs a convergence analysis using the BVPh 2.0 software, which involves 25 iterative cycles. The skin friction coefficient and Nusselt number are two key parameters. The results serve as a fundamental step for engineering applications of advanced fluid mechanics, which include the manufacture of materials, the transport of biomedicals, cooling systems, and thermal management.

DESIGNING AND IMPLEMENTING AN OPINION MINING ANALYSIS USING ARTIFICIAL INTELLIGENCE

2025-03-09T12:53:43+08:00

This study examines the impact of integrating, evaluating, executing, and analysing a model. This involves downloading Twitter information and inserting it into the MongoDB database. The Twitter samples and the extracted features, together with a trained classifier based on supervised learning, their polarity, and emotional words. The insights from the study will help in understanding sentiment analysis using machine learning techniques. The MongoDB database driver, data preprocessing, and sentiment analysis were successfully connected to retrieve text. Visualisations were successful. The application can display graphs and bar charts.

OVERVIEW OF LIVER FIBROSIS DETECTION METHOD USING MACHINE LEARNING APPROACHES

2025-09-05T10:59:09+08:00

Liver fibrosis is a chronic illness that results from chronic liver diseases such as hepatitis, cirrhosis, haemochromatosis, and non-alcoholic fatty liver disease (NAFLD). For timely detection and improved patient outcomes, liver fibrosis must be accurately staged for effective patient management and treatment. Traditional diagnostic methods such as liver biopsies, have risks and are invasive, among other downsides. However, recent advances in Machine Learning (ML) have offered substitutes to detect liver fibrosis. ML approaches are emerging as effective tools for the non-invasive detection of liver fibrosis; they have the potential to increase detection accuracy and reduce the demand for invasive liver biopsies. This overview provides a summary of methods for detecting liver fibrosis, with a particular emphasis on ML and both traditional and contemporary assessment methods. In recent years, many machine learning algorithms have been used to predict liver fibrosis detection for the Genetic Algorithm (GA), Artificial Neural Network (ANN), Naïve Bayes, and Multi-linear Regression, as well as Random Forest, Genetic techniques, Decision Tree (DT), Support Vector Machine (SVM), and Particle Swarm Optimisation (PSO). Besides, the results and performance of ML approaches are reviewed with a comparison of existing research studies, which were used for the detection of liver fibrosis. This review provides a roadmap that will assist researchers in making the most of the extensive capabilities of machine learning algorithms to build secure predictive models.