Classification by minimum-message-length inference

  • C. S. Wallace
Theory Of Computing, Algorithms And Programming
Part of the Lecture Notes in Computer Science book series (LNCS, volume 468)


Although classifiation is perhaps the oldest practical application of MML inference, the early algorithm was subject to weakly inconsistent estimation. The same problem is inherent in any MML inference which infers many discrete "nuisance" parameters. A solution has been found using a novel coding trick, which could be useful in many inductive inferences.


Classification Unsupervised learning Minimum message length Induction Coding Statistical inference 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Boulton, D.M. & Wallace, C.S. "A Program for Numerical Classification", Comp.J. 13,1, pp. 63–69, 1970Google Scholar
  2. Boulton, D.M. & Wallace, C.S. "An Information Measure for Hierarchic Classification", Comp.J. 16, 3, pp. 254–261, 1973Google Scholar
  3. Chaitin, G.J. "On the Length of Programs for Computing Finite Sequences", J.A.C.M. 13, 4, pp. 547–549, 1966Google Scholar
  4. Cheeseman,P.C. "AUTOCLASS II Conceptual Clustering System", Proc. Machine Learning Conference, pp. 54–64, 1988Google Scholar
  5. Chong,Y.H., Pham,B., Manton,M. & Maeder,A. "Automatic Nephanalysis from Infrared GMS Data", Proc. Australian Joint A.I. Conf, 1989Google Scholar
  6. Gath, I. & Geva, A.B. "Unsupervised Optimal Fuzzy Clustering", I.E.E.E. Trans. on Pattern Analysis and Machine Intelligence, 11, 7, pp. 773–781, 1989Google Scholar
  7. Georgeff, M.P. & Wallace, C.S. "A General Criterion for Inductive Inference", Proc. 6th European Conference on Artificial Intelligence, Tim O'Shea (ed.). Elsevier, Amsterdam, 1984Google Scholar
  8. Harman, H.H. "Modern Factor Analysis" (2nd ed.) Univ. of Chicago Press, Chicago, 1967Google Scholar
  9. Wallace, C.S. & Boulton, D.M. "An Information Measure for Classification", Comp.J., 11, pp. 185–195, 1968Google Scholar
  10. Wallace,C.S. "Inference and Estimation by Compact Coding", Monash University Computer Science Technical Report 46, 1984Google Scholar
  11. Wallace,C.S. & Patrick,J. "An Improved Program for Classification", Monash University Computer Science Technical Report 47, 1984Google Scholar
  12. Wallace, C.S. & Freeman, P.R. "Estimation and Inference by Compact Coding", J. R. Statist. Soc. B, 49, 3, pp. 240–265, 1987Google Scholar
  13. Wallace,C.S. & Freeman,P.R. "Single Factor Analysis by MML Estimation", Monash University Computer Science Technical Report 89/127, 1989Google Scholar
  14. Wallace,C.S. "False Oracles and SMML Estimation", Monash University Computer Science Technical Report 89/128, 1989Google Scholar
  15. Wallace,R.D. "Finding Natural Clusters through Entropy Minimization", CMU-CS-89-183, Computer Science, Carnegie Mellon University, 1989Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • C. S. Wallace
    • 1
  1. 1.Computer ScienceMonash UniversityClaytonAustralia

Personalised recommendations