Metode strojnog učenja u klasifikaciji signala EEG-a kod afektivnih poremećaja

undergraduate thesis

• Title (english): Machine Learning Methods for EEG Signal Classification in Affective Disorders
• Author: Ivan Skorupan
• Mentor: Mario Cifrek (mentor)
• Committee member: Mario Cifrek (predsjednik povjerenstva)
• Committee member: Krešimir Friganović (član povjerenstva)
• Committee member: Zrinka Topić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Nebalansirani skupovi podataka su česti u medicinskoj praksi. Klasični algoritmi strojnog učenja pate od prenaučenosti na nebalansiranim skupovima pa su potrebne posebne metode za unaprjeđenje modela. U ovom radu je provedena klasifikacija značajki signala EEG-a po medicinskim dijagnozama nad nekoliko tablica koristeći modele stroja potpornih vektora, k najbližih susjeda, slučajne šume te AdaBoosta. Analizirani su rezultati klasifikacije i pokazano je da modeli AdaBoost i stroj potpornih vektora s linearnom jezgrom pokazuju najveći potencijal za klasifikaciju and nebalansiranim skupovima podataka u većini primjera.

Abstract (english)

Imbalanced datasets are often found in medical praxis. Classical machine learning algorithms suffer from overfitting on imbalanced datasets, which is why special methods are needed to improve model performance. In this thesis, EEG signal features classification by medical diagnoses was performed on several tables using the support vector machine, k-nearest neighbors, random forest and AdaBoost models. Classification results were analyzed and it was shown that the AdaBoost and linear kernel support vector machine models show the most potential for classification on imbalanced datasets in majority of the examples.

• Keywords: EEG; elektroencefalografija; strojno učenje; nebalansirani skup podataka; afektivni poremećaj; stroj potpornih vektora; algoritam k najbližih susjeda; slučajna šuma; AdaBoost
• Keywords (english): EEG; electroencephalography; machine learning; imbalanced dataset; affective disorder; support vector machine; k-nearest neighbors algorithm; AdaBoost
• Language: english
• URN:NBN: urn:nbn:hr:168:077282
• 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: 2023-01-14 09:56:38