An Evaluation of the Usefulness of Case-Based Explanation

By Padraig Cunningham, Dónal Doyle, J. Loughrey. 2003

136 citations recorded by the source corpus.

One of the perceived benefits of Case-Based Reasoning (CBR) is the potential to use retrieved cases to explain predictions. Surprisingly, this aspect of CBR has not been much researched. There has been some early work on knowledge-intensive approaches to CBR where the cases contain explanation patterns (e.g. SWALE). However, a more knowledge-light approach where the case similarity is the basis for explanation has received little attention. To explore this, we have developed a CBR system for predicting blood-alcohol level. We compare explanations of predictions produced by this system with alternative rule-based explanations. The case-based explanations fare very well in this evaluation and score significantly better than the rule-based alternative.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

The Bayesian Case Model: A Generative Approach for Case-Based Reasoning and Prototype Classification (2014, background): “Studies show that example cases significantly improve user confidence in the resulting solutions, as compared to providing the solution alone or by also displaying a rule that was used to find the solution [11].”
Aspects of Intelligent Systems Explanation (2013, background): “Little empirical evidence is available comparing rule-based and case-based explanations, apart from the work completed by [40] who show that, from a relatively small sample, case-based are preferred to rule-based explanations.”
Interactive and interpretable machine learning models for human machine collaboration (2015, background): “Studies show that example cases significantly improve user confidence in the resulting solutions, as compared to providing the solution alone or by also displaying a rule that was used to find the solution [41].”
Introduction to the Special Issue on Explanation in Case-Based Reasoning (2005, background): “While rule-based explanations of an intelligent system’s conclusions remain an important legacy from expert systems research, recent research supports the view that an explanation based on experience may be more convincing than an explanation based on rules (Cunningham et al., 2003).”
Gaining insight through case-based explanation (2009, methodology): “It is useful to divide work on explanation in CBR into two categories (Cunningham et al., 2003).”
A Case-Based Explanation System for Black-Box Systems (2005, methodology): “Research by Cunningham et al. (2003) has further supported the claim that CBR explanations are more convincing than rule-based explanations in some domains.”
Explanation in Case-Based Reasoning–Perspectives and Goals (2005, background): “…and ad hoc reports that this works, there has been research showing that displaying cases along with the solution significantly improved user confidence in the solution compared to only showing the solution, or displaying a rule that was used in finding the solution (Cunningham et al., 2003).”
Explainable artificial intelligence for breast cancer: A visual case-based reasoning approach (2019, background): “It has been shown that explanations based on similar cases are more convincing than explanations based on rules [22], although this possibility has not been widely used and many explanation systems are not visual and rely on a single similar case.”
How Case-Based Reasoning Explains Neural Networks: A Theoretical Analysis of XAI Using Post-Hoc Explanation-by-Example from a Survey of ANN-CBR Twin-Systems (2019, background): “Finally, [73] did user tests to show that the retrieved cases have some explanatory value in these domains.”
A Review of Explanation and Explanation in Case-Based Reasoning (2003, background): “However some people argue that “knowledge intensive CBR is missing the point of CBR, which is the potential CBR has to finesse knowledge engineering effort by manipulating cases that are compiled chunks of knowledge” (Cunningham et al., 2003).”
Retrieval, reuse, revision and retention in case-based reasoning (2005, unknown): “1984; Ashley & Aleven, 1992; Leake, 1996b; McSherry, 2001b; Cunningham et al ., 2003; Doyle et al ., 2004; McSherry, 2004b; Leake & McSherry, 2005).”