Abstract
Mapping knowledge structures is a key task in Knowledge Discovery in Databases (KDD). In order to display the thematic organization of knowledge, we compare and evaluate two different cartography approaches: principal components analysis (PCA) and a multilayer perceptron (MLP) in “self-association” mode. This kind of MLP can be used to perform a PCA when the activation function is set to the identity function. This allows us to look for the non-linear activation function which best fits the data structure. We present an evaluation criterion and the results and maps obtained with both methods. We notice that the MLP detects a non-linearity in the data structure that the PCA does not detect. However, the MLP does not express the non-linearity completely. Finally we show how a related component analysis (RCA), based on graph theory, provides representations of the inter-clusters relationships, compensating for the approximate nature of the maps, and improving their readability.
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Bourret, P., Reggia, J., Samuelides, M.: Réseaux neuroneaux, une approche connexionniste de l’intelligence artificielle, Toulouse, TEKNEA (1991).
Brachman, R.J., Anand, T.: The Process of Knowledge Discovery in Databases, in Adavances in Knowledge Discovery and Data Mining. Fayyad, U.M., Piatetsky-Shapiro, G., Smyth, P., and Uthurusamy, R. Editors. Menlo Park, Calif. AAAI Press/The MIT Press, (1996) 37–57.
Elder, J., Pregribon, D.: A Statistical Perspective on Knowledge Discovery in Databases, in Databases, in Adavances in Knowledge Discovery and Data Mining. Fayyad, U.M., Piatetsky-Shapiro, G., Smyth, P., and Uthurusamy, R. Editors. Menlo Park, Calif. AAAI Press/The MIT Press (1996) 83–113.
Fayyad, U.M., Piatetsky-Shapiro, G., Smyth, P.: From Data Mining to Knowledge Discovery: An Overview, in Adavances in Knowledge Discovery and Data Mining. Fayyad, U.M., Piatetsky-Shapiro, G., Smyth, P., and Uthurusamy, R. Editors. Menlo Park, Calif. AAAI Press/The MIT Press (1996). 1–34.
Gallinari, P., Fogelman-Soulié, F.: Progressive Design of MLP Architecture”, Neuro-Nïmes (1988) 171–182.
Gallinari, P., Thiria, S., Fogelman-Soulié, F.: Multilayer perceptrons and data analysis, International Conference on neural networks, IEEE 1 (1988) 391–399.
Grivel, L., François, C.: Une station de travail pour classer, cartographier et anlyser l’information bibliographique dans une perspective de veille scientifique et technique, in Noyer, J.M. Editor, Les sciences de l’information: Bibliométrie, Scientométrie Informétrie, Rennes, SOLARIS II, Presses Universitaires de Rennes (1995) 81–112.
Holsheimer, M., Siebes, A.P. Data Mining: The Search for Knowledge in Databases. Amsterdam, CWI Report CS-R9406, ISSN 0169-118X. (1994).
Muller, Ch., Polanco, X., Royauté, J., Toussaint, Y.: Acquisition et structuration des connaissances en corpus: éléments méthodologiques. Rapport de recherche INRIA No 3198, available in postcript format ftp.inria.fr (192.93.2.54) (1997)
Polanco, X., Grivel, L., Royauté, J.: How to Do Things with Terms in Informetrics: Terminological Variation and Stabilization as Science Watch Indicators, Proceedins of the Fifth International Conference of the International Society for Scientometrics and Informetrics. Edited by M.E.D. Koening and A. Bookstein. Medford, NJ: Learned Information Inc (1995) 435–444.
Polanco, X.: Extraction et modélisation des connaissances: une approche et ses technologies, Premières Journées du Chapitre Français de l’International Society for Knowledge Organization (ISKO), à Lille, France, les 16–17 octobre 1997 (1997)
Polanco, X., François, C., Keim, J-P.: Artificial Neural Network Technology for the Classification and Cartography of Scientific and Technical Information, Scientometrics, vol. 41, no 1(1998), 69–82.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Polanco, X., François, C., Ould Louly, M.A. (1998). For visualization-based analysis tools in knowledge discovery process: A multilayer perceptron versus principal components analysis: A comparative study. In: Żytkow, J.M., Quafafou, M. (eds) Principles of Data Mining and Knowledge Discovery. PKDD 1998. Lecture Notes in Computer Science, vol 1510. Springer, Berlin, Heidelberg . https://doi.org/10.1007/BFb0094802
Download citation
DOI: https://doi.org/10.1007/BFb0094802
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65068-3
Online ISBN: 978-3-540-49687-8
eBook Packages: Springer Book Archive