SemLa: A Visual Analysis System for Fine-Grained Text Classification

Authors

  • Munkhtulga Battogtokh King's College London
  • Cosmin Davidescu ContactEngine
  • Michael Luck King's College London
  • Rita Borgo King's College London

DOI:

https://doi.org/10.1609/aaai.v38i21.30560

Keywords:

Artificial Intelligence, Natural language processing and speech recognition

Abstract

Fine-grained text classification requires models to distinguish between many fine-grained classes that are hard to tell apart. However, despite the increased risk of models relying on confounding features and predictions being especially difficult to interpret in this context, existing work on the interpretability of fine-grained text classification is severely limited. Therefore, we introduce our visual analysis system, SemLa, which incorporates novel visualization techniques that are tailored to this challenge. Our evaluation based on case studies and expert feedback shows that SemLa can be a powerful tool for identifying model weaknesses, making decisions about data annotation, and understanding the root cause of errors.

Downloads

Published

2024-03-24

How to Cite

Battogtokh, M., Davidescu, C., Luck, M., & Borgo, R. (2024). SemLa: A Visual Analysis System for Fine-Grained Text Classification. Proceedings of the AAAI Conference on Artificial Intelligence, 38(21), 23772-23774. https://doi.org/10.1609/aaai.v38i21.30560