Publication:
Integration Of Dynamic Loss Function Autoencoder In Boost

Simple item page
dc.contributor.authorShamsudin, Haziqah
dc.date.accessioned2026-05-11T02:21:22Z
dc.date.available2026-05-11T02:21:22Z
dc.date.issued2025-04
dc.description.abstractHighly class imbalance together with high data complexity (feature overlap and poor class separability), presents a significant challenge in machine learning. Traditional classifiers often exhibit bias towards the majority class, resulting in poor performance on the minority class, which is frequently the class of interest. Existing methods address imbalance or complexity, but rarely both effectively, and often lack adaptivity during training. This thesis addresses these challenges through a series of algorithmic enhancements.
dc.identifier.urihttps://erepo.usm.my/handle/123456789/24162
dc.language.isoen
dc.subjectMachine Learning
dc.titleIntegration Of Dynamic Loss Function Autoencoder In Boost
dc.typeResource Types::text::thesis::doctoral thesis
dspace.entity.typePublication
oairecerif.author.affiliationUniversiti Sains Malaysia
Files
Collections
Pusat Pengajian Sains Komputer - Tesis