RESEARCH ARTICLE Machine learning-based prediction of mechanical and thermal properties of nickel/cobalt/ferrous and dried leaves fiber-reinforced polymer hybrid composites H. Mohit 1 | M. R. Sanjay 2 | Suchart Siengchin 2 | Belal Kanaan 3 | Vakkar Ali 4 | Ibrahim M. Alarifi 4 | Tarek M. A. A. El-Bagory 4,5 1 Department of Mechanical Engineering, Alliance College of Engineering and Design, Alliance University, Bengaluru, India 2 Natural Composites Research Group Lab, Department of Materials and Production Engineering, The Siridorn International Thai-German Graduate School of Engineering (TGGS), King Mongkut's University of Technology North Bangkok (KMUTNB), Bangkok, Thailand 3 Department of Chemistry, College of Science in Zulfi, Majmaah University, Majmaah, Saudi Arabia 4 Department of Mechanical and Industrial Engineering, College of Engineering, Majmaah University, Riyadh, Saudi Arabia 5 Department of Mechanical Design, Faculty of Engineering Materia, Helwan University, Cairo, El-Mataria, Egypt Correspondence Ibrahim M. Alarifi, Department of Mechanical and Industrial Engineering, College of Engineering, Majmaah University, Al-Majmaah, 11952, Riyadh, Saudi Arabia. Email: i.alarifi@mu.edu.sa [Correction added on 11 October 2023, after first online publication: A few minor edits have been made in affiliations 2 and 5 in this version.] Abstract Dried leaves are the outstanding origin of cellulosic plant matter, and it is securing reputation as a renewable resource. Dried leaves fiber is suggested to possess the capability to substitute synthetic fibers in polymer laminates as a reinforcing component. The novelty of the present study reveals the effect of dried leaves fiber, cobalt, nickel, and ferrous reinforcement on the physical, mechanical, and thermal characteristics of epoxy, vinyl-ester, and polyester polymers using artificial neural network (ANN) technique. These composites were fabricated using ultrasonication bath-assisted wet layup method under ambient condition. The outcomes of this research exhibit that the dried leaves-cobalt fillers reinforced in all three polymers possess higher mechanical and thermal stability characteristics when compared with other samples. The rea- son may be assigned to producing novel hydroxyl functional groups and strong interfacial bonding of fillers within the matrix as observed from Fourier-transform infrared (FTIR) spectra and scanning electron microscope (SEM) micrographs, respectively. Moreover, as observed from the thermogravimetric analysis, the dried leaves-ferrous filler-reinforced polymer hybrid composites provided higher thermal stability. Statistical analysis was performed using the one-way ANOVA technique and found that outcomes were significant statistically under the confi- dence level of 95%. Hence, this investigation not only emphasize the significance of investigating new polymer composites but also highlight the benefits of engag- ing advanced modeling to forecast the material characteristics precisely. Highlights • Dried leaves and cobalt/nickel/ferrous are applied reinforcement to polymers. • Composites fabricated using ultrasonication bath-assisted wet layup technique. • LM Algorithm-based ANN selected for predicting the best composite. • Higher mechanical and thermal stability with dried leaves-cobalt filler. • One-way ANOVA proved statistically significant within the material properties. KEYWORDS fillers, hybrid composites, machine learning, natural fiber, polymer Received: 17 August 2023 Revised: 26 August 2023 Accepted: 17 September 2023 DOI: 10.1002/pc.27793 Polymer Composites. 2023;1–18. wileyonlinelibrary.com/journal/pc © 2023 Society of Plastics Engineers. 1