Active learning and its applications for reducing labeling effort in text classification tasks

Jacobs, Pieter (2021) Active learning and its applications for reducing labeling effort in text classification tasks. Bachelor's Thesis, Artificial Intelligence.

	Text Toestemming.pdf Restricted to Registered users only Download (120kB)
Preview	Text Bachelor_Thesis_Pieter_Jacobs.pdf Download (816kB) | Preview

Abstract

Labeling data can be an expensive task as it is usually performed manually by domain experts. This is cumbersome for deep learning, as it is dependent on large labeled datasets. Active learning (AL) is a paradigm that aims to reduce labeling effort by only using the data which the used model deems most informative. Little research has been done on AL in a text classification setting and next to none has involved the more recent, state-of-the-art NLP models. Here, we present an empirical study that compares different uncertainty-based algorithms with BERTbase as the used classifier. We evaluate the algorithms on two NLP classification datasets: Stanford Sentiment Treebank and KvK-Frontpages. Additionally, we explore heuristics that aim to solve presupposed problems of uncertainty-based AL. Namely, that it is unscalable and that it is prone to selecting outliers. Furthermore, we explore the influence of the query-pool size on the performance of AL. Our results show that using uncertainty-based AL with BERTbase outperforms randomly sampling data. This difference in performance can decrease as the query-pool size gets larger. It was also found that the proposed heuristics did not significantly improve performance when compared to the basic implementation of uncertainty-based AL.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Wiering, M.A. and Maillette de Buij Wenniger, G.E.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	10 Aug 2021 08:26
Last Modified:	18 Aug 2021 11:33
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/25626

Actions (login required)

View Item