Browsing by Author "Ahmadian, A."

Now showing 1 - 4 of 4

Citation Count: 0
Efficient solutions to time-fractional telegraph equations with Chebyshev neural networks
(Institute of Physics, 2024) Ali, A.H.; Senu, N.; Ahmadian, A.
This study aims to employ artificial neural networks (ANNs) as a novel method for solving time fractional telegraph equations (TFTEs), which are typically addressed using the Caputo fractional derivative in scientific investigations. By integrating Chebyshev polynomials as a substitute for the traditional hidden layer, computational performance is enhanced, and the range of input patterns is broadened. A feed-forward neural network (NN) model, optimized using the adaptive moment estimation (Adam) technique, is utilized to refine network parameters and minimize errors. Additionally, the Taylor series is applied to the activation function, which removes any limitation on taking fractional derivatives during the minimization process. Several benchmark problems are selected to evaluate the proposed method, and their numerical solutions are obtained. The results demonstrate the method’s effectiveness and accuracy, as evidenced by the close agreement between the numerical solutions and analytical solutions. © 2024 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.
Citation Count: 0
A promising exponentially-fitted two-derivative Runge-Kutta-Nystrom method for solving y′′ = f(x,y): Application to Verhulst logistic growth model
(Elsevier, 2024) Lee, K. C.; Nazar, R.; Senu, N.; Ahmadian, A.
Explicit exponentially-fitted two-derivative Runge-Kutta-Nystrom method with single f-function and multiple third derivatives is proposed for solving special type of second-order ordinary differential equations with exponential solutions. B-series and rooted tree theory for the proposed method are developed for the derivation of order conditions. Then, we build frequency-dependent coefficients for the proposed method by integrating the second-order initial value problem exactly with solution in the linear composition of set functions e(lambda t) and e(-lambda t) with lambda is an element of R. An exponentially-fitted two-derivative Runge-Kutta-Nystrom method with three stages fifth order is derived. Linear stability and stability region of the proposed method are analyzed. The numerical tests show that the proposed method is more effective than other existing methods with similar algebraic order in the integration of special type of second-order ordinary differential equations with exponential solutions. Also, the proposed method is used to solve a famous application problem, Verhulst logistic growth model and the result shows the proposed method still works effectively for solving this model.
Citation Count: 0
A systematic review on personalized hybrid diet recommendations
(Elsevier, 2024) Sandra, M.; Aarthi, K.; Thilagasree, C.S.; Ahmadian, A.; Narayanamoorthy, S.
A nutritious diet is necessary for a healthier life. A psychographic decision such as eating poor nutrient-dense food might have serious health effects on individuals. Inadequate eating habits are strongly associated with a higher risk of developing noncommunicable diseases, such as chronic respiratory, cardiovascular, diabetes, and hypertension. Nowadays, people focus more on taking pharmaceuticals than on changing their diets. For those with busy schedules and little awareness of healthy initiatives, an appropriate nutrition prescription is provided to help them avoid such repercussions. The purpose of the chapter is to analyze the multicriteria decision making (MCDM) tool's worldwide application and acceptance in relation to dietary guidelines. A thoughtful examination of 17 manuscripts published between 2015 and 2023 and catalogued by IEEE Xplore, Springer, Google Scholar, and Science Direct has been developed to comprehend the use of this tool and their function in nutritional requirements. This chapter advances to the corpus of knowledge on fuzzy MCDM by presenting a profound vision through a conceptual understanding of fuzzy MCDM in dietary suggestions from a mathematical perspective. According to the study's findings, traditional fuzzy sets were the most favored sets, and fuzzy analytic hierarchy process (AHP) was the most widely used MCDM approach, whether used independently or in conjunction with another method. Additionally, India has the most publications utilizing fuzzy MCDM techniques to address diet-related issues. © 2025 Elsevier Inc. All rights reserved.
Citation Count: 0
Uncertainty in Computational Intelligence-Based Decision Making: A volume in Advanced Studies in Complex Systems
(Elsevier, 2024) Ahmadian, A.; Salahshour, S.; Balas, V.E.; Baleanu, D.
Uncertainty in Computational Intelligence-Based Decision-Making focuses on techniques for reasoning and decision-making under uncertainty that are used to solve issues in artificial intelligence (AI). It covers a wide range of subjects, including knowledge acquisition and automated model construction, pattern recognition, machine learning, natural language processing, decision analysis, and decision support systems, among others. The first chapter of this book provides a thorough introduction to the topics of causation in Bayesian belief networks, applications of uncertainty, automated model construction and learning, graphic models for inference and decision making, and qualitative reasoning. The following chapters examine the fundamental models of computational techniques, computational modeling of biological and natural intelligent systems, including swarm intelligence, fuzzy systems, artificial neutral networks, artificial immune systems, and evolutionary computation. They also examine decision making and analysis, expert systems, and robotics in the context of artificial intelligence and computer science. © 2025 Elsevier Inc. All rights are reserved including those for text and data mining AI training and similar technologies.