Uvajanje metod strojnega učenja v študijske programe kemijskega inženirstva : magistrsko delo

Šuster, Jure

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Uvajanje metod strojnega učenja v študijske programe kemijskega inženirstva : magistrsko delo
Authors:	ID Šuster, Jure (Author) ID Novak-Pintarič, Zorka (Mentor) More about this mentor... ID Bogataj, Miloš (Comentor)
Files:	MAG_Suster_Jure_2025.pdf (3,02 MB) MD5: 319807BCA03253AE587FD4FCD69A6B82
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FKKT - Faculty of Chemistry and Chemical Engineering
Abstract:	V magistrskem delu so predstavljeni temeljni koncepti strojnega učenja in orodja za njihovo implementacijo, s posebnim poudarkom na praktičnem učenju z uporabo programskega jezika Python in njegovih uveljavljenih knjižnic, kot so NumPy, Pandas, Scikit-learn, Matplotlib, Keras in TensorFlow. Delo sistematično opisuje ključne faze razvoja modelov strojnega učenja, ki vključujejo pridobivanje in predobdelavo podatkov, izvedbo eksplorativne analize, uporabo različnih algoritmov za učenje ter vrednotenje njihove uspešnosti z ustreznimi metričnimi kazalci. V praktičnem delu so podrobneje prikazani trije primeri uporabe. Prvi primer obravnava napoved topnosti spojin, pri čemer je bil na osnovi javno dostopnega nabora podatkov razvit regresijski model za napoved kvantitativnih vrednosti. Drugi primer prikazuje uporabo klasifikacijskega modela za razvrščanje podatkov, pri čemer je bil poseben poudarek namenjen vrednotenju uspešnosti modela z različnimi metrikami. Tretji primer vključuje učenje večplastne nevronske mreže na znanem podatkovnem naboru ročno napisanih številk MNIST in ilustrira celoten proces – od priprave podatkov, nastavitve arhitekture modela, učenja modela, do vizualizacije in interpretacije rezultatov. Rezultati teh primerov potrjujejo, da je strojno učenje mogoče učinkovito približati študentom kemijskega inženirstva s kombinacijo teoretičnih osnov in praktičnih vaj, kar pomembno prispeva k njihovi boljši usposobljenosti za izzive sodobne industrije.
Keywords:	strojno učenje, kemijsko procesno inženirstvo, nevronske mreže, regresija, klasifikacija
Place of publishing:	Maribor
Place of performance:	Maribor
Publisher:	[J. Šuster]
Year of publishing:	2025
Number of pages:	1 spletni vir (1 datoteka PDF (X, 55 str.))
PID:	20.500.12556/DKUM-95754
UDC:	004.85:378.147(043.2)
COBISS.SI-ID:	255802115
Publication date in DKUM:	30.10.2025
Views:	169
Downloads:	46
Metadata:
Categories:	KTFMB - FKKT
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	19.10.2025

Secondary language

Language:	English
Title:	Incorporation of machine learning methods into chemical engineering study programs
Abstract:	This master’s thesis presents the fundamental concepts of machine learning and the tools required for their implementation, with a particular emphasis on practical learning through the use of the Python programming language and its widely adopted libraries, such as NumPy, Pandas, Scikit-learn, Matplotlib, Keras, and TensorFlow. The work systematically outlines the key stages involved in the development of machine learning models, which include data acquisition and preprocessing, conducting exploratory data analysis, applying various algorithms for learning, and evaluating their performance using appropriate metric indicators. In the practical section, three representative application examples are presented in detail. The first example focuses on predicting the solubility of compounds, where a regression model was developed to forecast quantitative values based on data obtained from a publicly available dataset. The second example illustrates the use of a classification model for data categorization, with particular attention given to assessing the performance of the model through different evaluation metrics. The third example involves training a multilayer neural network on the well-known MNIST dataset of handwritten digits, demonstrating the entire process—from preparing and preprocessing the data, configuring the model architecture, model training, to the visualization and interpretation of the results. The outcomes of these case studies confirm that machine learning can be effectively introduced to chemical engineering students by combining theoretical foundations with hands-on exercises, thereby enhancing their preparedness for the challenges of modern industry.
Keywords:	machine learning, chemical process engineering, neural networks, regression, classification

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back