-
Research Article
Image Clustering Using Exponential Regularized Discriminant Analysis
Issue:
Volume 12, Issue 1, February 2026
Pages:
1-12
Received:
16 September 2025
Accepted:
23 October 2025
Published:
26 January 2026
Abstract: When clustering images, the images are typically sampled as nonlinear manifolds. In this case, local learning-based image clustering models are used. Several proposed clustering models are based on linear discriminant analysis (LDA). In image clustering based on linear discriminant analysis (LDA), the problem of small-sample-size (SSS) is presented when the dimensionality of image data is larger than the number of samples. To solve this problem, various image clustering models based on local learning have been introduced. In the proposed clustering models, we added tuning parameters to deal with the small-sample-size (SSS) problem arising in linear discriminant analysis (LDA). In this paper, we propose an exponential regularized discriminant clustering model as an image clustering model based on local learning. In the proposed local exponentially regularized discriminant clustering (LERDC) model, the local scattering matrices of the regularized discriminant model are projected into the exponential domain to address the SSS problem of LDA. Compared with previous clustering methods based on local learning, k-nearest neighbors and regularization parameter λ in the local exponentially regularized discriminant clustering model are the tuning parameters for clustering. The experiments are concluded that the clustering performance of the proposed LERDC model is comparable to that of the clustering methods based on previous local learning.
Abstract: When clustering images, the images are typically sampled as nonlinear manifolds. In this case, local learning-based image clustering models are used. Several proposed clustering models are based on linear discriminant analysis (LDA). In image clustering based on linear discriminant analysis (LDA), the problem of small-sample-size (SSS) is presented...
Show More
-
Research Article
Application of Linear Programming to Optimize Production and Profit (A Case Study of Oda Natural Spring Water Factory)
Ahmed Buseri Ashine*
,
Mideksa Tola Jiru
Issue:
Volume 12, Issue 1, February 2026
Pages:
13-23
Received:
6 November 2025
Accepted:
14 January 2026
Published:
31 January 2026
Abstract: This research focuses on optimizing production processes at Oda Natural Spring Water Factory in Ethiopia through the application of Linear Programming (LP). In the competitive bottled water industry, efficient resource management is crucial for profitability. Oda Natural Spring Water, known for its unique selenium-rich composition, faces challenges related to inefficient production practices and suboptimal resource allocation. This study develops a tailored LP model to determine the optimal daily production quantities for four bottled water sizes (35cl, 60cl, 100cl, and 200cl), with the objective of maximizing profit while adhering to constraints such as production time, cost budget, and market demand limits. Data collected over six months were analyzed using Excel Solver. The findings indicate that the factory can achieve a maximum daily profit of 150,143 Ethiopian Birr (ETB) by producing 1,100 packs of 35cl, 1,714 packs of 60cl, 1,441 packs of 100cl, and 976 packs of 200cl water. Sensitivity analysis reveals that the production cost constraint is binding, while significant production time remains unused. The study underscores LP as a practical decision-making tool in manufacturing, providing actionable strategies for improving resource allocation, reducing costs, and enhancing profitability. Recommendations include cost reduction initiatives and regular review of production plans in response to market dynamics.
Abstract: This research focuses on optimizing production processes at Oda Natural Spring Water Factory in Ethiopia through the application of Linear Programming (LP). In the competitive bottled water industry, efficient resource management is crucial for profitability. Oda Natural Spring Water, known for its unique selenium-rich composition, faces challenges...
Show More
-
Review Article
A Review on Visco-Elastic Fluid Heat Transfer in Porous Media
Sridhar Kulkarni*,
Nagappa Kallappa Enagi
Issue:
Volume 12, Issue 1, February 2026
Pages:
24-30
Received:
14 December 2025
Accepted:
30 December 2025
Published:
4 February 2026
DOI:
10.11648/j.ijtam.20261201.13
Downloads:
Views:
Abstract: This review presents a stimulating discussion in which, in a systematic and coherent way, the governing equations and the different mathematical models used in the characterization of the heat transfer properties of viscoelastic fluids flowing through porous structures are illuminated. Particular attention is paid to the dissection of the role played by several key parameters that control the flow and thermal regime. These are the basic porosity effects, which govern the available pathways for flow and interfacial area; the form of the drag forces experienced, normally modelled via extensions of Darcy’s Law, such as the Darcy-Brinkman-Forchheimer formulation, accounting for the effects of viscous diffusion and inertial resistance; and the influence of the boundary effects, which often introduce nonlinear velocity and temperature gradients near the confining walls. Moreover, a significant portion of the research is devoted to the details of fluid rheology, studying how various viscoelastic constitutive models, such as Oldroyd-B, Maxwell, or generalized power-law models, interact with the geometric constraints imposed by the porous matrix to modify momentum and energy transport. Important research findings indicate ways in which non-Newtonian behavior and non-equilibrium conditions can profoundly impact the more traditional predictions of heat transfer. For example, the elasticity of the fluid may either stabilize or destabilize the onset of thermal convection and lead to novel flow patterns and heat transport mechanisms absent in simpler Newtonian-fluid systems. These important findings are synthesized within the review to provide the researcher and engineer with a consolidated view of acquired knowledge and quantitative insight. Beyond its consolidation of current knowledge, this comprehensive review undertakes a critical review of the current state-of-the-art. It painstakingly identifies persistent gaps in existing literature, highlighting outstanding areas where understanding at either the theoretical or experimental level remains incomplete. The result is the formulation of specific high-priority, potential future research directions. The suggested avenues for investigation are often novel rheological models, micro-scale pore-level heat transfer mechanisms, or coupled phenomena in non-idealized porous structures, opening new avenues for further development of the fundamental knowledge base and practical uses of heat transport phenomena in complex viscoelastic-porous systems.
Abstract: This review presents a stimulating discussion in which, in a systematic and coherent way, the governing equations and the different mathematical models used in the characterization of the heat transfer properties of viscoelastic fluids flowing through porous structures are illuminated. Particular attention is paid to the dissection of the role play...
Show More
-
Research Article
The Existence and Uniqueness of SDE Solutions Diriged by Multiractional Brownian Motion , Based on Several Developed Methods
Hamid EBeye,
Moussa Thioune,
Demba Bocar Ba*
Issue:
Volume 12, Issue 1, February 2026
Pages:
31-36
Received:
30 November 2025
Accepted:
5 January 2026
Published:
4 February 2026
DOI:
10.11648/j.ijtam.20261201.14
Downloads:
Views:
Abstract: To overcome some of the limitations of traditional fractional Brownian motion (fBm), multifractional Brownian motion (mBm) was created. The Holder exponent of mBm can vary along the trajectory, unlike fBm, which is advantageous for modeling processes whose regularity varies over time, like internet traffic or photos. This is the main difference between the two processes. Many continuous observations can be modeled by stochastic differential equations governed by mBm, especially in biology and finance. Using a non-stationary multifractional Brownian motion with a time-dependent Hurst parameter as the noise source and a simply linear drift coefficient, we first show the existence and uniqueness of a solution to stochastic differential equations in this article. Next, we examine multifractional Brownian motion (mBm) driven stochastic differential equation models that allow for the simulation of some discontinuity-containing phenomena.
Abstract: To overcome some of the limitations of traditional fractional Brownian motion (fBm), multifractional Brownian motion (mBm) was created. The Holder exponent of mBm can vary along the trajectory, unlike fBm, which is advantageous for modeling processes whose regularity varies over time, like internet traffic or photos. This is the main difference bet...
Show More
