Skip to main content
SpringerLink
Log in

Interpretation and knowledge discovery from the multilayer perceptron network: Opening the black box

  • Articles
  • Published:
Neural Computing & Applications Aims and scope Submit manuscript

Abstract

This paper interprets the outputs from the multilayer perceptron (MLP) network by finding the input data features at the input layer of the network which activate the hidden layer feature detectors. This leads directly to the deduction of the significant data inputs, the inputs that the network actually uses to perform the input/output mapping for a classification task, and the discovery of the most significant of these data inputs. The analysis presents a method for providing explanations for the network outputs and for representing the knowledge learned by the network in the form of significant input data relationships. During network development the explanation facilities and data relationships can be used for network validation and verification, and after development, for rule induction and data mining where this method provides a potential tool for knowledge discovery in databases (KDD).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Rumelhart DE, Hinton GE, Williams RJ. Learning Internal representations by Error Propagation. In: Rumelhart DE, McClelland JL and PDP Research Group (eds), Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Vol. 1: Foundations. MIT Press, Cambridge, MA, 1986

    Google Scholar 

  2. Werbos P. Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. PhD diss. Harvard University, Cambridge, MA, 1974

    Google Scholar 

  3. Lisboa PGJ. Neural Networks Current Applications. Chapman & Hall, London, 1992

    Google Scholar 

  4. Dayhoff J. Neural Network Architectures. An Introduction. Van Nostrand Reinhold, New York, 1990

    Google Scholar 

  5. Spiegelhalter DJ, Taylor CC (eds). Machine Learning, Neural and Statistical Classification. Ellis Horwood, Chichester, 1994

    Google Scholar 

  6. Vaughn ML. An Introduction to Neural Computing. International CIS Journal 1989; 3(2).

  7. The International Journal of Neural Networks Research & Applications 1989; 1(1) (Learned Information (Europe) Ltd, Oxford, England)

  8. Hecht-Nielsen R. Neurocomputing. Addison-Wesley, Wokingham, UK, 1990

    Google Scholar 

  9. Wasserman PD. Neural Computing Theory and Practice. Van Nostrand Reinhold, New York, 1989

    Google Scholar 

  10. Cotter F. An Investigation into the Properties of Artificial Neural Network Architectures that Achieve Speaker Recognition. BSc(Hons) Command and Control, Communications and Information Systems, Project Report. Cranfield University, RMCS, 1992

  11. Best Practice Guidelines for Developing Neural Computing Applications. DTI, London, 1994

  12. Lippmann RP. An Introduction to Computing with Neural Nets. IEEE ASSP Magazine 1987; April: 4–22

    Google Scholar 

  13. Cassell I. An Investigation into Neural Computing as an Approach to Modelling Life Assurance Under-writing. BSc(Hons) Command and Control, Communications and Information Systems, Project Report, Cranfield University, RMCS, 1991

  14. Refenes AN, Azema-Barac M. Neural Networks in Financial Asset Management. Neural Computing & Applications 1994; 2(1)

  15. Khabaza T, Shearer C. Data Mining with Clementine. IEE Colloquium, Knowledge Discovery in Databases. London, February 1995

  16. le Cun Y. Generalisation and Network Design Strategies. Technical Report CRG-TR-89-4, University of Toronto, Department of Computer Science, 1989

  17. Cavill SJ. The Use of Artificial Neural Networks in the Assessment and Diagnosis of Lower Back Pain. MSc Project Report, Cranfield University, RMCS, 1994

  18. Vaughn ML, Van Schalkwyk H, King RA. The Application of Artificial Neural Networks to Provide Safe Access to Driver Information Systems and Other Non-critical Automotive Functions. 8th Int Conf “Automotive Electronics” IEE, London, 1991

  19. Campbell MI. Speaker Independent Phoneme Recognition from the TIMIT Database using TESPAR. A comparison of the MLP and LVQ Artificial Neural networks as Classifiers. MSc Project Report, Cranfield University, RMCS, 1994

  20. Lison AC. Using Neural Networks for the Prediction of Helicopter Airframe Load Spectra from Flight Parameter Data. MSc Project Report, Cranfield University, RMCS, 1994

Download references

Author information

Authors and Affiliations

  1. Computing Information Systems Management Group, Cranfield University, RMCS, SN6 8LA, Shrivenham, Swindon, UK

    Marilyn Lougher Vaughn

Authors
  1. Marilyn Lougher Vaughn
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Vaughn, M.L. Interpretation and knowledge discovery from the multilayer perceptron network: Opening the black box. Neural Comput & Applic 4, 72–82 (1996). https://doi.org/10.1007/BF01413743

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF01413743

Keywords

Search

Navigation