Implementacija i analiza algoritma analize glavnih komponenti

undergraduate thesis

• Title (english): Implementation and analysis of the principal component analysis algorithm
• Author: Antonio Lukić
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: David Dukić (član povjerenstva)
• Committee member: Josip Jukić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rastuća količina podataka koja se generira iz godine u godinu zahtijeva učinkovite metode za obradu i analizu. U ovom radu implementiran je i istražen algoritam analize glavnih komponenti. Opisane su matematičke metode te model autoenkodera potrebnog za razumijevanje algoritma. Implementacija algoritma ostvarena je na dva načina: maksimizacijom varijance u prostoru projekcije te minimizacijom rekonstrukcijske pogreške. Uspoređeni su rezultati dviju implementacija nad stvarnim i umjetnim podacima te su prikazane prednosti i mane svake od implementacija. Završno, uspoređena je učinkovitost implementacija s postojećim rješenjima.

Abstract (english)

The growing amount of data generated year after year requires efficient methods for processing and analysis. In this work, the principle component analysis algorithm was implemented and researched. Mathematical methods and the autoencoder model were described to understand the algorithm. The algorithm was implemented in two ways: maximization of the variance in the projection space and minimization of the reconstruction error. The results of two implementations on real and artificial data were compared, and the advantages and disadvantages of each implementation were presented. Finally, the effectiveness of the implementations was compared with already implemented solutions.

• Keywords: strojno učenje; neuronske mreže; redukcija dimenzionalnosti; PCA; linearni autoenkoder; svojstvena vrijednost
• Keywords (english): machine learning; neural network; dimensionality reduction; PCA; linear autoencoder; eigenvalue
• Language: croatian
• URN:NBN: urn:nbn:hr:168:726374
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-07-13 10:17:07