Istraživanje temperatura taljenja i kristalnih struktura metala primjenom strojnog učenja

master's thesis

• Title (english): Study of melting temperatures and crystal structures of metals using machine learning
• Author: Borna Jelić
• Mentor: Goranka Bilalbegović (mentor)
• Committee member: Goranka Bilalbegović (predsjednik povjerenstva)
• Committee member: Ivan Kupčić (član povjerenstva)
• Committee member: Maja Planinić (član povjerenstva)
• Committee member: Mario Basletić (član povjerenstva)
• Committee member: Danko Radić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Physics); Zagreb
• Defense date and country: 2021-03-26, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Physics

Abstract

Strojno učenje postaje sve značajnije za znanost i gospodarstvo. U ovom diplomskom radu se za analizu osobina elementarnih metala koriste metode nadziranog strojnog učenja linearna regresija i neuronske mreže. Metodom linearne regresije analizirane su korelacije temperatura taljenja za Youngove module elastičnosti, konstante rešetke i koeficijente linearne termičke ekspanzije. Primjenom neuronskih mreža ispitana je mogućnost predviđanja strukture kristalne rešetke metala. U metodičkom poglavlju predstavljeno je upoznavanje sa strojnim učenjem za učenike četvrtog razreda srednjih škola i to za k-sredine, algoritam nenadziranog strojnog učenja. Korišten je programski jezik Python i njegove biblioteke za strojno učenje.

Abstract (english)

Machine learning is becoming increasingly important for science and economics. In this diploma thesis, the methods of supervised machine learning, linear regression and neural networks, are used to analyze the properties of elemental metals. The correlations of melting temperatures for Young ’s modulus of elasticity, lattice constants and coefficients of linear thermal expansion were analyzed by the linear regression method. The possibility of predicting the structure of the metal crystal lattice was investigated using neural networks. The methodical section presents the introduction to machine learning for fourth grade high school students using k-means, an algorithm of unsupervised machine learning. The Python programming language and its machine learning libraries were used.

• Keywords: fizika materijala; metali; nadzirano strojno učenje; linearna regresija; neuronske mreže; strojno učenje u srednjim školama; nenadzirano strojno učenje; algoritam k-sredina
• Language: croatian
• URN:NBN: urn:nbn:hr:217:132676
• Study programme: Title: Physics and Computer Science; specializations in: Education; Course: Education; Study programme type: university; Study level: integrated undergraduate and graduate; Academic / professional title: magistar/magistra fizike i informatike (magistar/magistra fizike i informatike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2021-04-28 11:29:20