A Parameter Matrix Based Approach to Computing Minimal Hitting Sets

Conference paper
Part of the Studies in Computational Intelligence book series (SCI, volume 431)

Abstract

Computing all minimal hitting sets is one of the key steps in model-based diagnosis. Because of the low capabilities due to the expansion of state space in large-scale system diagnosis, more efficient approximation algorithms are in motivation. A matrix-based minimal hitting set (M-MHS) algorithm is proposed in this paper. A parameter matrix records the relationships between elements and sets and the initial problem is divided into several sub-problems by decomposition. The efficient prune rules avoid the computation of the sub-problems without solutions. Parameterized way and de-parameterized way are both given so that the more suitable algorithm could be chosen according to the cases. The simulation results show that, the proposed algorithm outperforms HSSE and BNB-HSSE in large-scale problems and keeps a relatively stable performance when data changes in different regulations. The algorithm provides a valuable tool for computing hitting sets in model-based diagnosis of large-scale systems.

Keywords

minimal hitting set model-based diagnosis parameter matrix 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    de Kleer, J., Williams, B.C.: Diagnosing multiple faults. Artificial Intelligence 32(1), 97–130 (1987)MATHCrossRefGoogle Scholar
  2. 2.
    Williams, B.C., Ragno, R.J.: Conflict-directed A* and its role in model-based embedded systems. Discrete Applied Mathematics 155(12), 1562–1595 (2007)MathSciNetMATHCrossRefGoogle Scholar
  3. 3.
    Reiter, R.: A theory of diagnosis from first principles. Artificial Intelligence 32(1), 57–95 (1987)MathSciNetMATHCrossRefGoogle Scholar
  4. 4.
    Greiner, R., Smith, B.A., Wilkerson, R.W.: A correction to the algorithm in reiter’s theory of diagnosis. Artificial Intelligence 41(1), 79–88 (1989)MathSciNetMATHCrossRefGoogle Scholar
  5. 5.
    Wotawa, F.: A variant of Reiter’s hitting-set algorithm. Information Processing Letters 79(1), 45–51 (2001)MathSciNetMATHCrossRefGoogle Scholar
  6. 6.
    Yunfei, J., Li, L.: Computing the Minimal Hitting Sets with Binary HS-Tree. Journal of Software 13(12), 2267–2274 (2002) (in Chinese)Google Scholar
  7. 7.
    Yunfei, J., Li, L.: The Computing of Hitting Sets with Boolean Formulas. Chinese Journal of Computers 26(8), 919–924 (2003) (in Chinese)Google Scholar
  8. 8.
    Jie, H., Lin, C., Peng, Z.: A Compounded Genetic and Simulated Annealing Algorithm for Computing Minimal Diagnosis. Journal of Software 15(9), 1345–1350 (2004) (in Chinese)Google Scholar
  9. 9.
    Li, L., Jiang, Y.: Computing minimal hitting sets with genetic algorithm. In: Proceedings of the 13th International Workshop on Principles of Diagnosis, Austria, pp. 77–80 (2002)Google Scholar
  10. 10.
    Zhao, X., Ouyang, D.: A method of combining SE-tree to compute all minimal hitting sets. Progress in Natural Science 16(2), 169–174 (2006)MathSciNetMATHCrossRefGoogle Scholar
  11. 11.
    Chen, X., Meng, X., Qiao, R.: Method of computing all minimal hitting set based on BNB-HSSE. Chinese Journal of Scientific Instrument 31(1), 61–67 (2010) (in Chinese)Google Scholar
  12. 12.
    Li, L.: Computing minimal hitting sets with logic array in model-based diagnosis. Journal of Jinan University (Natural Science) 23(1), 24–27 (2002) (in Chinese)Google Scholar
  13. 13.
    Fijany, A., Vatan, F.: New Approaches for Efficient Solution of Hitting Set Problem. In: ACM International Conference Proceeding Series (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Dong Wang
    • 1
  • Wenquan Feng
    • 1
  • Jingwen Li
    • 1
  • Meng Zhang
    • 1
  1. 1.School of Electronics and Information EngineeringBeijing University of Aeronautics and AstronauticsBeijingChina

Personalised recommendations