Abstract
Large multi-label datasets contain labels that occur thousands of times (frequent group), those that occur only a few times (few-shot group), and labels that never appear in the training dataset (zero-shot group). Multi-label few- and zero-shot label prediction is mostly unexplored on datasets with large label spaces, especially for text classification. In this paper, we perform a fine-grained evaluation to understand how state-of-the-art methods perform on infrequent labels. Furthermore, we develop few- and zero-shot methods for multi-label text classification when there is a known structure over the label space, and evaluate them on two publicly available medical text datasets: MIMIC II and MIMIC III. For few-shot labels we achieve improvements of 6.2% and 4.8% in R@10 for MIMIC II and MIMIC III, respectively, over prior efforts; the corresponding R@10 improvements for zero-shot labels are 17.3% and 19%.
Original language | English |
---|---|
Title of host publication | Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 |
Editors | Ellen Riloff, David Chiang, Julia Hockenmaier, Jun'ichi Tsujii |
Pages | 3132-3142 |
Number of pages | 11 |
ISBN (Electronic) | 9781948087841 |
State | Published - 2018 |
Event | 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 - Brussels, Belgium Duration: Oct 31 2018 → Nov 4 2018 |
Publication series
Name | Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 |
---|
Conference
Conference | 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 |
---|---|
Country/Territory | Belgium |
City | Brussels |
Period | 10/31/18 → 11/4/18 |
Bibliographical note
Publisher Copyright:© 2018 Association for Computational Linguistics
ASJC Scopus subject areas
- Computational Theory and Mathematics
- Computer Science Applications
- Information Systems