Artificial Intelligence Review

, Volume 12, Issue 1–3, pp 95–115 | Cite as

Artificial Keys for Botanical Identification using a Multilayer Perceptron Neural Network (MLP)

  • Jonthan Y. Clark
  • Kevin Warwick


In this paper, practical generation of identification keys for biological taxa using a multilayer perceptron neural network is described. Unlike conventional expert systems, this method does not require an expert for key generation, but is merely based on recordings of observed character states. Like a human taxonomist, its judgement is based on experience, and it is therefore capable of generalized identification of taxa. An initial study involving identification of three species of Iris with greater than 90% confidence is presented here. In addition, the horticulturally significant genus Lithops (Aizoaceae/Mesembryanthemaceae), popular with enthusiasts of succulent plants, is used as a more practical example, because of the difficulty of generation of a conventional key to species, and the existence of a relatively recent monograph. It is demonstrated that such an Artificial Neural Network Key (ANNKEY) can identify more than half (52.9%) of the species in this genus, after training with representative data, even though data for one character is completely missing.

Aizoaceae expert system identification Iris key Lithops Mesembryanthemaceae multilayer perceptrons neural network 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Clark, J. Y. (1996). A key to Lithops N. E. Br. (Aizoaceae). Bradleya 14: 1–9.Google Scholar
  2. Clark, J. Y. & Warwick, K. (1995). Detection of faults in a high speed packaging machine using a multilayer perceptron (MLP). IEE Colloquium: Innovations in manufacturing control through mechatronics, Newport, Gwent, UK, Digest No. 95/214: 7/1–7/3.Google Scholar
  3. Cole, D. T. (1986). Lithops Locality Data. Desmond T. Cole: Swakaroo, Emmarentia, South Africa, January.Google Scholar
  4. Cole, D. T. (1988). Lithops, Flowering Stones. Randburg, South Africa: Acorn Books.Google Scholar
  5. Dallwitz, M. J. (1974). A flexible computer program for generating identification keys. Systematic Zoology 23: 50–57.Google Scholar
  6. Dallwitz, M. J. (1980). A general system for coding taxonomic descriptions. Taxon 29: 41–46.Google Scholar
  7. Dallwitz, M. J., Paine, T. A. & Zurcher, E. J. (1993). User's guide to the DELTA system: a general system for processing taxonomic descriptions, 4th edition. Canberra, Australia: CSIRO Division of Entomology.Google Scholar
  8. DeBoer, H. W. & Boom, B. K. (1964). An analytical key for the genus Lithops. National Cactus & Succulent Society Journal 19: 34–37, 51–55.Google Scholar
  9. Everitt, B. S. (1993). Cluster Analysis. New York: Edward Arnold/Halsted Press.Google Scholar
  10. Fearn, B. (1981). Lithops. Oxford, UK: British Cactus & Succulent Society (Handbook No. 4).Google Scholar
  11. Fisher, R. A. (1936). The use of multiple measurements in taxonomic problems. Annals of Eugenics 7: 179–188.Google Scholar
  12. Goodacre, R. (1994). Characterisation and quantification of microbial systems using pyrolysis mass spectrometry: introducing neural networks to analytical pyrolysis. Microbiology Europe 2(2): 16–22.Google Scholar
  13. Goodacre, R., Kell, D. B. & Bianchi, G. (1992). Neural networks and olive oil. Nature 359: 594.Google Scholar
  14. Goodacre, R., Trew, S., WrigleyJones, C., Neal, M. J., Maddock, J., Ottley, T. W., Porter, N. & Kell, D. B. (1994a). Rapid screening for metabolite overproduction in fermentor broths, using pyrolysis mass spectrometry with multivariate calibration and artificial neural networks. Biotechnology and Bioengineering 44: 1205–1216.Google Scholar
  15. Goodacre, R., Neal, M. J., Kell, D. B., Greenham, L. W., Nobel W. C. & Harvey, R. G. D. (1994b). Rapid identification using pyrolysis mass spectrometry and artificial neural networks of Propionibacterium acnes isolated from dogs. Journal of Applied Bacteriology 76: 124–134.Google Scholar
  16. Hammer, S. A. & Uijs, R. (1994). Lithops coleorum S. A. Hammer & R. Uijs sp. nov., a new species of Lithops N. E. Br. from the Northern Transvaal. Aloe 31(2): 36–38.Google Scholar
  17. Haykin, S. (1994). Neural networks – a comprehensive foundation. New York: Macmillan College Publishing Company, Inc.Google Scholar
  18. Lobanov, A. J., Schilow, W. F. & Nikritin, L. M. (1981). Zur Anwendung von Computern für die Determination in der Entomologie. Deutsche Entomologie Zeitung 28: 29–43.Google Scholar
  19. Mathew, B. (1981). The Iris. London: B. T. Batsford Ltd.Google Scholar
  20. Matthews, C. P., Clark, J. Y., Sharkey, P. M. & Warwick, K. (1995). A comparison of cluster analysis and neural networks for the reliability of machinery. Proceedings SPIE Conference Photons East. Philadelphia.Google Scholar
  21. Pankhurst, R. J. (1991). Practical Taxonomic Computing. UK: University of Cambridge Press.Google Scholar
  22. Pankhurst, R. J. & Aitchison, R. R. (1975). A computer program to construct polyclaves. In Pankhurst, R. J. (ed.) Biological Identification with Computers, 73–78. London and Orlando: Academic Press.Google Scholar
  23. Partridge, T. R., Dallwitz, M. J. & Watson, L. (1993). A primer for the DELTA system, 3rd edition. Canberra, Australia: CSIRO Division of Entomology.Google Scholar
  24. Ray, A. K. (1991). Equipment fault diagnosis – A neural network approach. Computers in Industry 16: 169–177.Google Scholar
  25. Rumelhart, D. E. & McClelland, J. L. (1986). Parallel Distributed Processing, Vols. 1 & 2. Cambridge, Mass.: MIT Press.Google Scholar
  26. Wallace, R. S. (1990). Systematic significance of allozyme variation in the genus Lithops (Mesembryanthemaceae). Mitt. Inst. Allg. Bot. Hamburg: Proceedings of the twelfth plenary meeting of aetfat. Symposium VI, 509–524. Hamburg, Germany: Band 23b.Google Scholar
  27. Yoon, Y., Brobst, R. W., Bergstresser, P. R. & Peterson, L. (1989). A desktop neural network for dermatology diagnosis. Journal of Neural Network Computing 1: 43–52.Google Scholar

Copyright information

© Kluwer Academic Publishers 1998

Authors and Affiliations

  • Jonthan Y. Clark
    • 1
  • Kevin Warwick
    • 1
  1. 1.Department of CyberneticsUniversity of ReadingWhiteknights, ReadingUK

Personalised recommendations