Browsing by Author "Kunrada Kankam"

Now showing 1 - 4 of 4
  • Default Image
    Item
    A VARIANT OF THE PROXIMAL GRADIENT METHOD FOR CONSTRAINED CONVEX MINIMIZATION PROBLEMS
    (Mathematical Research Press, 2024) Suparat Kesornprom; Kunrada Kankam; Papatsara Inkrong; Nattawut Pholasa; Prasit Cholamjiak; P. Cholamjiak; School of Science, University of Phayao, Phayao, 56000, Thailand; email: prasit.ch@up.ac.th
    This paper presents a new variant of the proximal gradient algorithm based on double inertial extrapolation to solve a constrained convex minimization problem in real Hilbert spaces. We discuss its weak convergence, including numerical image and signal recovery experiments to support the main results. Some comparisons with other algorithms are also provided. The experiments demonstrate that our method converges better than the other methods in the literature. ©2024 Journal of Nonlinear Functional Analysis.
  • Default Image
    Item
    An inertial projective forward-backward-forward algorithm for constrained convex minimization problems and application to cardiovascular disease prediction
    (International Scientific Research Publications, 2024) Prasit Cholamjiak; Watcharaporn Cholamjiak; Kunrada Kankam; P. Cholamjiak; School of Science, University of Phayao, Phayao, 56000, Thailand; email: prasit.ch@up.ac.th; W. Cholamjiak; School of Science, University of Phayao, Phayao, 56000, Thailand; email: watcharaporn.ch@up.ac.th; K. Kankam; Elementary Education Program, Faculty of Education, Suan Dusit University Lampang Center, Lampang, 52100, Thailand; email: kunradazzz@gmail.com
    In this paper, we introduce a novel machine learning algorithm designed for the classification of cardiovascular diseases. The proposed inertial projected forward-backward-forward algorithm is developed to address constrained minimization in Hilbert spaces, with a specific focus on improving the accuracy of disease classification. Utilizing inertial techniques, the algorithm employs a projected forward-backward-forward strategy, demonstrating convergence under mild conditions. Evaluation of the algorithm employs four essential performance metrics-accuracy, F1-score, recall, and precision to gauge its effectiveness compared to alternative classification models. Results indicate significant performance gains, achieving peak metrics of 77.50% accuracy, 71.57% precision, 91.27% recall, and 80.23% F1-score, thereby surpassing established benchmarks in machine learning models for cardiovascular disease classification. © 2025 All rights reserved.
  • Default Image
    Item
    Inertial iterative method for solving equilibrium problems and fixed point problems
    (Springer Nature, 2024) Min Li; Zhongbing Xie; Prasit Cholamjiak; Kunrada Kankam; Z. Xie; School of Mathematics and Statistics, Wuhan University, Wuhan, 430072, China; email: xzbmath@163.com
    In this paper, we present an inertial iterative method for solving pseudomonotone equilibrium and fixed point problems in Banach spaces. Under appropriate conditions, we improve the convergence efficiency of our proposed algorithm by introducing a new step size and iteration rule, and further derive a strong convergence theorem. Finally, we demonstrate through numerical experiments that our new algorithm compares favourably with existing methods in terms of convergence behaviour. © 2024, The Author(s) under exclusive licence to Sociedade Brasileira de Matem‡tica Aplicada e Computacional.
  • Default Image
    Item
    Three-step projected forwardÐbackward algorithms for constrained minimization problem
    (Springer Nature, 2025) Kunrada Kankam; Muhammad Aslam Noor; Prasit Cholamjiak; P. Cholamjiak; School of Science, University of Phayao, Phayao, 56000, Thailand; email: prasitch2008@yahoo.com
    We design new projective forwardÐbackward algorithms for constrained minimization problems. We then discuss its weak convergence via a new linesearch that the hypothesis on the Lipschitz constant of the gradient of functions is avoided. We provide its applications to solve image deblurring and image inpainting. Finally, we discuss the optimal selection of parameters that are proposed in algorithms in terms of PSNR and SSIM. It reveals that our new algorithm outperforms some recent methods introduced in the literature. © The Author(s) under exclusive licence to Korean Society for Informatics and Computational Applied Mathematics 2024.