prikaz prve stranice dokumenta Klasifikacija bolesti na temelju podataka o ekspresiji gena
No public access
undergraduate thesis
Klasifikacija bolesti na temelju podataka o ekspresiji gena
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2023. urn:nbn:hr:168:226316
Tuković, Lana
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Tuković, L. (2023). Klasifikacija bolesti na temelju podataka o ekspresiji gena (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:226316

Tuković, Lana. "Klasifikacija bolesti na temelju podataka o ekspresiji gena." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2023. https://urn.nsk.hr/urn:nbn:hr:168:226316

Tuković, Lana. "Klasifikacija bolesti na temelju podataka o ekspresiji gena." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2023. https://urn.nsk.hr/urn:nbn:hr:168:226316

Tuković, L. (2023). 'Klasifikacija bolesti na temelju podataka o ekspresiji gena', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 26 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:226316

Tuković L. Klasifikacija bolesti na temelju podataka o ekspresiji gena [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2023 [cited 2024 December 26] Available at: https://urn.nsk.hr/urn:nbn:hr:168:226316

L. Tuković, "Klasifikacija bolesti na temelju podataka o ekspresiji gena", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:168:226316

Metadata
TitleKlasifikacija bolesti na temelju podataka o ekspresiji gena
Title (english)Disease classification from gene expression data
AuthorLana Tuković
MentorKrešimir Križanović (mentor)
Committee memberKrešimir Križanović (predsjednik povjerenstva)
Committee memberJosipa Lipovac (član povjerenstva)
Committee memberRafael Josip Penić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2023-07-06, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Ekspresija gena odnosi se na proces kojim se informacije kodirane u DNA pretvaraju u funkcionalne molekule poput proteina. Razvojem tehnika analize ekspresije gena u posljednjim godinama omogućen je uvid u uzroke i liječenje različitih bolesti, uključujući karcinom. Cilj rada bio je razviti modele strojnog učenja za klasifikaciju karcinoma primijenjujući tehnike nadziranog i nenadziranog strojnog učenja, kao i duboko učenje, te usporediti njihove performanse. Implementirani su algoritmi logističke regresije, stroja potpornih vektora, dubokog učenja i grupiranja uz pomoć programskih biblioteka sklearn i keras. Za postizanje boljih rezultata kod modela dubokog učenja istražili smo tehniku poznatu kao Analiza glavnih komponenti (PCA) i model ranog zaustavljanja te usporedili rezultate. Algoritme smo primijenili na skupu podataka za klasifikaciju leukemije i skupu podataka za klasifikaciju raka dojke. Uspješno smo klasificirali pojednine karcinome modelima strojnog učenja i dobili rezultate u skladu s očekivanjima.
Abstract (english)
Gene expression refers to the process by which information encoded in DNA is transformed into functional molecules such as proteins. The development of gene expression analysis techniques in recent years has provided insights into the causes and treatment of various diseases, including cancer. The aim of this study was to develop machine learning models for cancer classification by applying supervised and unsupervised learning techniques, as well as deep learning, and to compare their performances. The implemented algorithms included logistic regression, support vector machine, deep learning, and clustering using the sklearn and keras programming libraries. To improve the results of the deep learning models, we experimented with a technique known as Principal Component Analysis (PCA) and early stopping, and compared the results. We applied the algorithms to datasets intended for leukemia and breast cancer classification. Using machine learning models and obtained results in line with expectations, we successfully classified different types of cancer.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:226316
Study programmeTitle: 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 resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-07-13 10:37:36