Enumerating Minimal Explanations by Minimal Hitting Set Computation

We consider the problem of enumerating minimal explanations in propositional theory. We propose a new way of characterizing the enumeration problem in terms of not only the number of explanations, but also the number of unexplanations. Maximal unexplanations are a maximal set of abducible formulas which cannot explain the observation given a background theory. In this paper, we interleavingly enumerate not only minimal explanations but also maximal unexplanations. To best of our knowledge, there has been no algorithm which is characterized in terms of such maximal unexplanations. We propose two algorithms to perform this task and also analyze them in terms of query complexity, space complexity and time complexity.

[1]  Raymond Reiter,et al.  A Theory of Diagnosis from First Principles , 1986, Artif. Intell..

[2]  Dimitrios Gunopulos,et al.  Data mining, hypergraph transversals, and machine learning (extended abstract) , 1997, PODS.

[3]  Kazuhisa Makino,et al.  Generating All Abductive Explanations for Queries on Propositional Horn Theories , 2003, CSL.

[4]  Kazuhisa Makino,et al.  Abduction and the Dualization Problem , 2003, Discovery Science.

[5]  Antonis C. Kakas,et al.  The role of abduction in logic programming , 1998 .

[6]  Georg Gottlob,et al.  The complexity of logic-based abduction , 1993, JACM.

[7]  Luigi Portinale,et al.  On the role of abduction , 1995, CSUR.

[8]  Leonid Khachiyan,et al.  On the Complexity of Dualization of Monotone Disjunctive Normal Forms , 1996, J. Algorithms.

[9]  Takeaki Uno,et al.  Detailed Description of an Algorithm for Enumeration of Maximal Frequent Sets with Irredundant Dualization , 2003, FIMI.

[10]  Kazuhisa Makino,et al.  On computing all abductive explanations , 2002, AAAI/IAAI.

[11]  Takeaki Uno,et al.  Enumerating Maximal Frequent Sets Using Irredundant Dualization , 2003, Discovery Science.

[12]  James Bailey,et al.  Discovery of Minimal Unsatisfiable Subsets of Constraints Using Hitting Set Dualization , 2005, PADL.

[13]  Dimitrios Gunopulos,et al.  Data mining, hypergraph transversals, and machine learning (extended abstract) , 1997, PODS '97.

[14]  Heikki Mannila,et al.  On an algorithm for finding all interesting sentences , 1996 .

[15]  Bart Selman,et al.  Horn Approximations of Empirical Data , 1995, Artif. Intell..

[16]  Ian Witten,et al.  Data Mining , 2000 .

[17]  H. Mannila,et al.  Discovering all most specific sentences , 2003, TODS.