Strojno učenje na nebalansiranim skupovima podataka

master's thesis

• Title (english): Machine learning on imbalanced datasets
• Author: Silvija Grgić
• Mentor: Boris Vrdoljak (mentor)
• Committee member: Boris Vrdoljak (predsjednik povjerenstva)
• Committee member: Mario Brčić (član povjerenstva)
• Committee member: Mirta Baranović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Skupovi podataka koje koriste algoritmi strojnog učenja mogu biti nebalansirani i ta
karakteristika može utjecati na rezultate tih algoritama. U sklopu ovog diplomskog rada
opisani su pojedini algoritmi strojnog učenja i dva nebalansirana skupa podataka. Prikazani
su rezultati algoritama strojnog učenja i neuronske mreže nad tim nebalansiranim skupovima
podataka ali i rezultati istih algoritama i neuronske mreže nad istim skupovima podataka
samo uz primjenu SMOTE i NearMiss metoda za balansiranje nebalansiranih skupova
podataka.

Abstract (english)

Data sets used by machine learning algorithms can be unbalanced and this characteristic can
affect the results of these algorithms. As a part of this thesis, few machine learning
algorithms and two unbalanced data sets are described. The results of machine learning and
neural network algorithms over these unbalanced data sets are presented, but also the results
of the same algorithms and neural network over the same data sets only with the application
of SMOTE and NearMiss methods for balancing unbalanced data sets.

• Keywords: strojno učenje; nebalansirani skupovi podataka; SMOTE; NearMiss; neuronske mreže
• Keywords (english): machine learning; imbalanced datasets; SMOTE; NearMiss; neural networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:389445
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 12:21:46