Concept description vectors and the 20 question game.

Wlodzislaw Duch1,2, Julian Szymański2, and Tomasz Sarnatowicz2.
1School of Computer Engineering, Nanyang Technological University, Singapore,
and 2Department of Informatics, Nicolaus Copernicus University,
Grudziadzka 5, 87-100 Torun, Poland.


Knowledge of properties that are applicable to a given object is a necessary prerequisite to formulate intelligent question. Concept description vectors provide simplest representation of this knowledge, storing for each object information about the values of its properties. Experiments with automatic creation of concept description vectors from various sources, including ontologies, dictionaries, encyclopedias and unstructured text sources, are described. Information collected in this way is used to formulate questions that have high discriminating power in the twenty questions game.

Preprint for comments in PDF, 98 KB.

Reference: Intelligent Information Processing and Web Mining, Advances in Soft Computing, Springer Verlag, ISBN 3-540-25056-5 (Eds. Klopotek, M.A., Wierzchon, S.T., Trojanowski, K.) (2005) 41-50

BACK to the publications of W. Duch.
BACK to the on-line publications of the Department of Informatics, NCU.