An Interactive Tool for Interpretability of Time Series Classification

Abstract

In this demo paper we put forward an explainable artificial intelligence (XAI) tool for model-agnostic instance-based interpretability of time series classification. Our tool allows for model inspection on user-generated instances, together with generation and visualization of counterfactuals. This will boost the understanding of black box classification models, offering the user the ability to navigate the boundary between classes.

Cite

Text

Håvardstun et al. "An Interactive Tool for Interpretability of Time Series Classification." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2024. doi:10.1007/978-3-031-70371-3_28

Markdown

[Håvardstun et al. "An Interactive Tool for Interpretability of Time Series Classification." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2024.](https://mlanthology.org/ecmlpkdd/2024/havardstun2024ecmlpkdd-interactive/) doi:10.1007/978-3-031-70371-3_28

BibTeX

@inproceedings{havardstun2024ecmlpkdd-interactive,
  title     = {{An Interactive Tool for Interpretability of Time Series Classification}},
  author    = {Håvardstun, Brigt Arve Toppe and Ferri, Cèsar and Telle, Jan Arne},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2024},
  pages     = {399-403},
  doi       = {10.1007/978-3-031-70371-3_28},
  url       = {https://mlanthology.org/ecmlpkdd/2024/havardstun2024ecmlpkdd-interactive/}
}