CoChat
Sign Up
Cited by 2,148

The DOI matches its Crossref record.

Using mutual information for selecting features in supervised neural net learning

Battiti, R.·1994·IEEE Transactions on Neural Networks
DOI 10.1109/72.298224
Open this paper in CoChat — free account

Abstract

This paper investigates the application of the mutual information criterion to evaluate a set of candidate features and to select an informative subset to be used as input data for a neural network classifier. Because the mutual information measures arbitrary dependencies between random variables, it is suitable for assessing the "information content" of features in complex classification tasks, where methods bases on linear relations (like the correlation) are prone to mistakes. The fact that the mutual information is independent of the coordinates chosen permits a robust estimation. Nonetheless, the use of the mutual information for tasks characterized by high input dimensionality requires suitable approximations because of the prohibitive demands on computation and samples. An algorithm is proposed that is based on a "greedy" selection of the features and that takes both the mutual information with respect to the output class and with respect to the already-selected features into account. Finally the results of a series of experiments are discussed.

Overview

Integrity signals

Registry

Crossref · 10.1109/72.298224

Cites

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

No references could be resolved yet.

Research works printed but not linked: 23

  1. 10.1109/72.80236 · DOI
  2. 10.1109/5.58325 · DOI · Paper page
  3. li, 1990, Mutual information functions versus correlation functions, J Stat Phys, 60, 823, 10.1007/BF01025996 · DOI
  4. 10.1162/neco.1989.1.3.402 · DOI
  5. 10.1109/72.80306 · DOI
  6. etti-spaccamela, 1990, Learning Boolean formulae using decision trees, Proc 3rd Workshop on Parallel Architectures and Neural Networks, 797
  7. mozer, 1989, Advances in neural information processing systems, 1, 107
  8. 10.1023/A:1022611825350 · DOI
  9. pernot, 1991, Determining the relevant parameters for the classification on a multi-layer perceptron: Application to radar data, Proc 1991 Int Conf Artificial Neural Networks ICANN-91, 797
  10. rumelhart, 1987, Parallel Distributed Processing Explorations in the Microstructure of Cognition, 1, 318
  11. 10.1111/j.1469-1809.1936.tb02137.x · DOI · Paper page
  12. duda, 1973, Pattern Classification and Scene Analysis
  13. 10.1103/PhysRevA.33.1134 · DOI
  14. 10.1016/0167-2789(89)90263-7 · DOI
  15. 10.1016/0893-6080(88)90023-8 · DOI
  16. 10.1016/0893-6080(89)90003-8 · DOI · Paper page
  17. bridle, 1990, Advances in neural information processing systems, 2, 211
  18. 10.1016/0893-6080(89)90030-0 · DOI
  19. 10.1109/18.87006 · DOI
  20. shannon, 1949, The Mathematical Theory of Communication
  21. teukolsky, 1988, Numerical Recipes in C
  22. weigend, 1991, Advances in neural information processing systems, 3, 875
  23. zemel, 1991, Advances in neural information processing systems, 3, 299

Cited by

Sorted by citations

Metadata

Authors
Battiti, R.
Venue
IEEE Transactions on Neural Networks
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
1994-07
Volume
5
Issue
4
Pages
537-550
ISSN
1045-9227
DOI
10.1109/72.298224
Cited by
2,148
References
0
Licence
ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html (vor)

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review