Strojno učenje za inženirje: Koncepti, primeri in uporaba v okolju MATLAB

Authors

Janez Gotlih
University of Maribor, Faculty of Mechanical Engineering
https://orcid.org/0000-0002-3853-8522
Miran Brezočnik
University of Maribor, Faculty of Mechanical Engineering
DOI: https://doi.org/10.18690/um.fs.10.2025

Keywords:

machine learning, supervised learning, unsupervised learning, reinforcement learning, transfer learning, MATLAB, engineering applications

Synopsis

Machine Learning for Engineers: Concepts, examples, and applications in MATLAB. The book deala with machine learning from the perspective of its application in engineering, linking fundamental concepts with practical application in the MATLAB environment. Four basic approaches to machine learning are presented: supervised learning, unsupervised learning, reinforcement learning, and transfer learning. For each approach, basic concepts, specific use cases, and independent work assignments are provided. Special emphasis is placed on the use of tools such as Regression Learner, Classification Learner, Deep Network Designer, and Reinforcement Learning Designer, with which students develop models based on data derived from real engineering examples. These include tool wear, machine vibrations, system balancing, and object recognition. The scripts also include experimental data sets and practical guidelines for learning, validating, and improving models. They are intended for students of technical disciplines and anyone who wants to learn how to use machine learning methods to solve specific engineering problems.

Downloads

Download data is not yet available.

Author Biographies

Janez Gotlih, University of Maribor, Faculty of Mechanical Engineering

Janez Gotlih is an assistant professor at the Faculty of Mechanical Engineering, University of Maribor. His professional work focuses on research in intelligent manufacturing systems, covering manufacturing technologies, assembly, robotics, and the application of optimization methods and machine learning. He is the author of numerous scientific, technical, and conference papers and an active contributor to national and international research and development projects. As a guest editor and reviewer, he collaborates with renowned international scientific journals. He integrates his research with industrial practice and the educational process, bringing the latest knowledge in machine learning and engineering optimization into real-world and academic applications.

Maribor, Slovenia. E-mail: janez.gotlih@um.si

Miran Brezočnik, University of Maribor, Faculty of Mechanical Engineering

Miran Brezočnik is a Full Professor at the Faculty of Mechanical Engineering, University of Maribor. He is the author of numerous original scientific articles, most of them published in journals indexed in the Web of Science database, as well as several university textbooks, lecture notes, and scientific monographs. He serves as the Editor-in-Chief of the international scientific journal Advances in Production Engineering & Management and is a member of the editorial boards of several distinguished international scientific publications. His scientific and professional work focuses primarily on modern manufacturing technologies, production modeling and optimization, automation, assembly, artificial intelligence, and machine learning.

Maribor, Slovenia. E-mail: miran.brezocnik@um.si

References

Bishop, C. M., & Nasrabadi, N. M. (2006). Pattern recognition and machine learning. Springer.

Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning. MIT press.

Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning. Springer.

Kuhn, M., & Johnson, K. (2016). Applied predictive modeling. Springer.

MathWorks. Statistics and Machine Learning Toolbox Documentation. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/stats/

MathWorks. Deep Learning Toolbox Documentation. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/deeplearning/

MathWorks. Reinforcement Learning Toolbox Documentation. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/reinforcement-learning/

MathWorks. Transfer Learning with Deep Network Designer. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/deeplearning/ug/transfer-learning.html

MathWorks. Build Condition Model for Industrial Machinery and Manufacturing Processes. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/stats/build-condition-model-for-industrial-machinery-and-manufacturing-processes.html

MathWorks. Three-Axis Vibration Data Set for Anomaly Detection. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/predmaint/anomaly-detection.html

MathWorks. (2024). Design and Train Agent Using Reinforcement Learning Designer. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/reinforcement-learning/ug/design-dqn-using-rl-designer.html

MathWorks. Train DQN Agent to Balance Cart-Pole System. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/reinforcement-learning/ug/train-dqn-agent-to-balance-cart-pole-system.html

MathWorks. Retrain Neural Network to Classify New Images. Dostopno 1. 10. 2025 na: https://www.mathworks.com/help/deeplearning/ug/retrain-neural-network-to-classify-new-images.html

Wikipedia. Confusion matrix. Dostopno 1. 10. 2025 na: https://en.wikipedia.org/wiki/Confusion_matrix

Wikipedia. Principal component analysis. Dostopno 1. 10. 2025 na: https://en.wikipedia.org/wiki/Principal_component_analysis

Downloads

E-book (pdf)

Published

November 10, 2025

Categories

Copyright (c) 2025 University of Maribor, University of Maribor Press

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Details about this monograph

THEMA Subject Codes (93)

U, UY

ISBN-13 (15)

978-961-299-078-7

COBISS.SI ID (00)

256145411

Date of first publication (11)

2025-11-10

How to Cite

Gotlih, J., & Brezočnik, M. (2025). Strojno učenje za inženirje: Koncepti, primeri in uporaba v okolju MATLAB. University of Maribor Press. https://doi.org/10.18690/um.fs.10.2025