University of Twente Student Theses


Cold-start Active Learning for Text Classification of Business Documents

Bachir Kaddis Beshay, Amir (2023) Cold-start Active Learning for Text Classification of Business Documents.

[img] PDF
Abstract:The thesis centers on two pivotal phases of active learning: the ’cold start’ and the subsequent ’warm start’. The quality of the initial pool of labeled data, often referred to as the ’cold start’ phase, significantly influences the efficiency and accuracy of ensuing learning iterations. However, this critical phase remains underexplored, particularly in the context of text classification. The study aims to bridge this knowledge gap, focusing on techniques that can judiciously construct an initial labeled pool to enable more effective sampling decisions in later iterations, ultimately optimizing the active learning process.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Interaction Technology MSc (60030)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page