A Case-Based Explanation System for Black-Box Systems

By Conor Nugent, Padraig Cunningham. 2005

96 citations recorded by the source corpus.

Most users of machine-learning products are reluctant to use them without any sense of the underlying logic that has led to the system’s predictions. Unfortunately many of these systems lack any transparency in the way they operate and are deemed to be black boxes. In this paper we present a Case-Based Reasoning (CBR) solution to providing supporting explanations of black-box systems. This CBR solution has two key facets; it uses local information to assess the importance of each feature and using this, it selects the cases from the data used to build the black-box system for use in explanation. The retrieval mechanism takes advantage of the derived feature importance information to help select cases that are a better reflection of the black-box solution and thus more convincing explanations.

Subfields: Artificial Intelligence

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

Citations in the corpus

The Twin-System Approach as One Generic Solution for XAI: An Overview of ANN-CBR Twins for Explaining Deep Learning (2019, methodology): “The historical work on twin systems divides into a late-1990s collection of papers by a Korean group at KAIST (e.g., [Park et al., 2004]) and some mid-2000s work by an Irish CBR group (e.g., [Cunningham et al., 2003; Nugent and Cunningham, 2005; Doyle et al., 2004]).”
'It's Reducing a Human Being to a Percentage': Perceptions of Justice in Algorithmic Decisions (2018, background): “1 Case-based explanations provide examples from the model’s training data which most closely resemble the output in question [12, 54], while ‘demographic’ explanations provide statistics on the outcome classes for relevant demographic variables [2].”
Meaningful Explanations of Black Box AI Decision Systems (2019, background): “…are model-dependent and aim, e.g., at explaining the decisions of neural networks by means of saliency masks, i.e., the portions of the input record (such as the regions of an image) that are mainly responsible for the outcome (Xu et al. 2015; Zhou et al. 2016; Nugent and Cunningham 2005).”
Explanation in Case-Based Reasoning–Perspectives and Goals (2005, methodology): “This method may even be used to provide explanations for nonCBR systems, as demonstrated by Nugent and Cunningham (2005).”
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): “The Irish researchers, who did not cite the Korean work, proposed a new and intriguing local feature-weighting method [74-76].”
Gaining insight through case-based explanation (2009, methodology): “The framework was developed to provide case-based explanations for black-box systems because of the strengths of this form of explanation (Nugent and Cunningham, 2004).”
Neural network Trojan (2013, unknown): “It is reasonable to assume that since the neural network is considered by most people as a black box [37], that contraband can be inserted into the neural network in the form of information in the weights and programming that become the neural Trojan.”
Why and why not explanations improve the intelligibility of context-aware intelligent systems (2009, background): “Among the myriad of machine learning algorithms, decision trees and Naïve Bayes lend themselves to be more explainable and transparent, while others are black-box algorithms that are not readily interpretable (e.g., Support Vector Machines and Neural Networks) [22].”