Towards feasible counterfactual explanations: a taxonomy guided template-based NLG method.

Salimi, Pedram; Wiratunga, Nirmalie; Corsar, David; Wijekoon, Anjana

Abstract

Counterfactual Explanations (cf-XAI) describe the smallest changes in feature values necessary to change an outcome from one class to another. However, many cf-XAI methods neglect the feasibility of those changes. In this paper, we introduce a novel approach for presenting cf-XAI in natural language (Natural-XAI), giving careful consideration to actionable and comprehensible aspects while remaining cognizant of immutability and ethical concerns. We present three contributions to this endeavor. Firstly, through a user study, we identify two types of themes present in cf-XAI composed by humans: content-related, focusing on how features and their values are included from both the counterfactual and the query perspectives; and structure-related, focusing on the structure and terminology used for describing necessary value changes. Secondly, we introduce a feature actionability taxonomy with four clearly defined categories, to streamline the explanation presentation process. Using insights from the user study and our taxonomy, we created a generalisable template-based natural language generation (NLG) method compatible with existing explainers like DICE, NICE, and DisCERN, to produce counterfactuals that address the aforementioned limitations of existing approaches. Finally, we conducted a second user study to assess the performance of our taxonomy-guided NLG templates on three domains. Our findings show that the taxonomy-guided Natural-XAI approach (n-XAIT) received higher user ratings across all dimensions, with significantly improved results in the majority of the domains assessed for articulation, acceptability, feasibility, and sensitivity dimensions.

Citation

SALIMI, P., WIRATUNGA, N., CORSAR, D. and WIJEKOON, A. 2023. Towards feasible counterfactual explanations: a taxonomy guided template-based NLG method. In Gal, K., Nowé, A., Nalepa, G.J., Fairstein, R. and Rădulescu, R. (eds.) ECAI 2023: proceedings of the 26th European conference on artificial intelligence (ECAI 2023), 30 September - 4 October 2023, Kraków, Poland. Frontiers in artificial intelligence and applications, 372. Amsterdam: IOS Press [online], pages 2057-2064. Available from: https://doi.org/10.3233/FAIA230499

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 26th European conference on artificial intelligence 2023 (ECAI-2023)
• Start Date: Sep 30, 2023
• End Date: Oct 5, 2023
• Acceptance Date: Jul 15, 2023
• Online Publication Date: Sep 28, 2023
• Publication Date: Dec 31, 2023
• Deposit Date: Jul 20, 2023
• Publicly Available Date: Aug 8, 2023
• Publisher: IOS Press
• Peer Reviewed: Peer Reviewed
• Pages: 2057-2064
• Series Title: Frontiers in artificial intelligence and applications
• Series Number: 372
• Series ISSN: 0922-6389; 1879-8314
• Book Title: ECAI 2023: proceedings of the 26th European conference on artificial intelligence (ECAI 2023), 30 September - 4 October 2023, Kraków, Poland
• ISBN: 9781643684369
• DOI: https://doi.org/10.3233/faia230499
• Keywords: Counterfactual explanation; Natural-XAI method; Feature actionability taxonomy (FAT); Actionability knowledge
• Public URL: https://rgu-repository.worktribe.com/output/2015280