Age Prediction from Skeletal Indicators using Computational Intelligence Methods. Zdenek Buk, Pavel Kordik, Miroslav Snorek. IWIM, Prague, 2007.
Article (in pdf)
Abstract. This paper presents the work of age prediction of human beings from their skeletal indicators using computational intelligence methods, such as feedforward neural networks, learning vector quantization (LVQ) and group of adaptive models evolution (GAME). The anthropological data set we have performed our experiments on, contains a lot of noise, which is characteristic feature of almost all data collected by observations. Goal of this work is to get the best possible results of the age prediction on such noisy data and to compare the results of particular methods.
Keywords. Prediction, anthropology, real noisy data, neural networks, GAME, inductive modelling.
References.
- T. Kohonen, J. Hynninen, J. Kangas, J. Laaksonen, K. Torkkola. LVQ PAK: The Learning Vector Quantization Program Package. Laboratory of Computer and Information Science. Helsinki University of Technology. Finland. 1995.
- P. Kord´ık.: Fully Automated Knowledge Extraction using Group of Adaptive Models Evolution. PhD thesis, Czech Technical University in Prague, FEE, Department of Computer Science and Engineering, FEE, CTU Prague, Czech Republic, September 2006.
- Ian H.Witten, Eibe Frank. DataMining?: PracticalMachine? Learning Tools and Techniques. 2005. ISBN 0-12-088407-0. Online version: http://www.cs.waikato.ac.nz/˜ml/weka/
