论文信息 - A Characterization of Contrastive Explanations Computation

A Characterization of Contrastive Explanations Computation

Abduction in artificial intelligence in a narrow sense is a methodology for finding explanation to a question in a deductive model. The problem of selecting which explanation is better suited to explaining a question is still unresolved. This paper proposes a computational model of explanation selection based on the concept of contrastive explanation popularly known in literature of inference to the best explanation. The proposed model of explanation finding for the question why P is replaced by finding the explanation for the question why P instead of Q given a particular contrast Q.

Alex Kean

[1] Patrick Suppes,et al. Introduction To Logic , 1958 .

[2] Alex Kean,et al. ASSUMPTION‐BASED REASONING AND CLAUSE MANAGEMENT SYSTEMS , 1990, Comput. Intell..

[3] G. Harman. The Inference to the Best Explanation , 1965 .

[4] Raymond Reiter,et al. Foundations of Assumption-based Truth Maintenance Systems: Preliminary Report , 1987, AAAI.

[5] Aristotle,et al. The Works of Aristotle , 1929, Nature.

[6] Peter Lipton,et al. Inference to the best explanation , 1993 .

[7] Alex Kean,et al. CLAUSE MANAGEMENT SYSTEMS (CMS) , 1993, Comput. Intell..