Skip to main content

Research Repository

Advanced Search

Generating Textual Explanations for Machine Learning Models Performance: A Table-to-Text Task

Ampomah, Isaac; Burton, James; Enshaei, Amir; Al Moubayed, Noura

Generating Textual Explanations for Machine Learning Models Performance: A Table-to-Text Task Thumbnail


Authors

Isaac Ampomah

James Burton james.burton@durham.ac.uk
PGR Student Doctor of Philosophy

Amir Enshaei



Contributors

Calzolari Nicoletta
Editor

Bechet Frederic
Editor

Blache Philippe
Editor

Choukri Khalid
Editor

Cieri Christopher
Editor

Declerck Thierry
Editor

Goggi Sara
Editor

Isahara Hitoshi
Editor

Maegaard Bente
Editor

Mariani Joseph
Editor

Mazo Helene
Editor

Odijk Jan
Editor

Piperidis Stelios
Editor

Abstract

Numerical tables are widely employed to communicate or report the classification performance of machine learning (ML) models with respect to a set of evaluation metrics. For non-experts, domain knowledge is required to fully understand and interpret the information presented by numerical tables. This paper proposes a new natural language generation (NLG) task where neural models are trained to generate textual explanations, analytically describing the classification performance of ML models based on the metrics’ scores reported in the tables. Presenting the generated texts along with the numerical tables will allow for a better understanding of the classification performance of ML models. We constructed a dataset comprising numerical tables paired with their corresponding textual explanations written by experts to facilitate this NLG task. Experiments on the dataset are conducted by fine-tuning pre-trained language models (T5 and BART) to generate analytical textual explanations conditioned on the information in the tables. Furthermore, we propose a neural module, Metrics Processing Unit (MPU), to improve the performance of the baselines in terms of correctly verbalising the information in the corresponding table. Evaluation and analysis conducted indicate, that exploring pre-trained models for data-to-text generation leads to better generalisation performance and can produce high-quality textual explanations.

Citation

Ampomah, I., Burton, J., Enshaei, A., & Al Moubayed, N. (2022). Generating Textual Explanations for Machine Learning Models Performance: A Table-to-Text Task. In C. Nicoletta, B. Frederic, B. Philippe, C. Khalid, C. Christopher, D. Thierry, …P. Stelios (Eds.),

Conference Name 13th Conference on Language Resources and Evaluation (LREC 2022)
Conference Location Marseille, France
Start Date Jun 20, 2022
End Date Jun 25, 2022
Online Publication Date Jun 20, 2022
Publication Date 2022
Deposit Date Apr 26, 2022
Publicly Available Date Jun 26, 2022
Pages 3542-3551
Public URL https://durham-repository.worktribe.com/output/1137326
Publisher URL http://www.lrec-conf.org/proceedings/lrec2022/index.html

Files








You might also like



Downloadable Citations