This paper addresses the problem of learning concept descriptions that are interpretable, or explainable. Explainability is understood as the ability to justify the learned concept in terms of the existing background knowledge. The starting point for the work was an existing system that would induce only fully explainable rules. The system performed well when the model used during induction was complete and correct. In practice, however, models are likely to be incomplete and incorrect. We report here a new approach that achieves explainability with models that are incomplete or inconsistent, or both. The basis of the system is the standard inductive search driven by an accuracy-oriented heuristic, biased towards rule explainability. The bias is abandoned when there is heuristic evidence that a significant loss of accuracy results from constraining the search to explainable rules only. The user can express their relative preference for accuracy vs. explainability. Experiments with the system indicate that, even with a partially incomplete and/or incorrect model, insisting on explainability results in only a small loss of accuracy. We also show how the new approach described can repair a faulty model using evidence derived from data during induction. Explainable Induction with an Imperfect Qualitative Model 2
[1]
Alberto Maria Segre,et al.
Programs for Machine Learning
,
1994
.
[2]
Stan Matwin,et al.
Using Qualitative Models to Guide Inductive Learning
,
1993,
ICML.
[3]
Ivan Bratko,et al.
KARDIO - a study in deep and qualitative knowledge for expert systems
,
1989
.
[4]
Douglas H. Fisher,et al.
Overcoming process delays with decision tree induction
,
1994,
IEEE Expert.
[5]
Peter Clark,et al.
Rule Induction with CN2: Some Recent Improvements
,
1991,
EWSL.
[6]
Kenneth D. Forbus.
Qualitative Process Theory
,
1984,
Artif. Intell..
[7]
Nada Lavrac,et al.
The Multi-Purpose Incremental Learning System AQ15 and Its Testing Application to Three Medical Domains
,
1986,
AAAI.
[8]
Peter Struss,et al.
Qualitative Reasoning
,
1997,
The Computer Science and Engineering Handbook.
[9]
Ivan Bratko,et al.
Machine Learning: Between Accuracy and Interpretability
,
1997
.
[10]
Stan Matwin,et al.
Learning Domain Theories using Abstract Beckground Knowledge
,
1993,
ECML.
[11]
Julio Ortega,et al.
Flexibly Exploiting Prior Knowledge in Empirical Learning
,
1995,
IJCAI.