When a Bayesian classifier is designed, a model for the class probability density functions (PDFs) has to be chosen. This choice is determined by a trade-off between robustness and low complexity -- which is usually satisfied by simple parametric models, based on a restricted number of parameters -- and the model's ability to fit a large class of PDFs -- which usually requires a high number of model parameters. In this paper, a model is introduced, where the class PDFs are approximated as piecewise multi-linear functions (a generalisation of bilinear functions for an arbitrary dimensionality). This model is compared with classical parametric and non-parametric models, from a point of view of versatility, robustness and complexity. The results of classification and PDF estimation experiments are discussed.
Nyssen, E, Van Kempen, L & Sahli, H 2000, "Pattern Classiffcation Based on a Piecewise Multi-Linear Model for the Class Probability Densities. in SSPR 2000, SPR 2000, Proc. Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recog- nition and Statistical Pattern Recognition; Alicante, Spain; August 30 - September 1, 2000.. The Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recognition (SSPR 2000) and Statistical Pattern Recognition (SPR 2000), pp. 501-510, Alicante, Spain., pp. 501-510, Unknown, 1/01/00.
Nyssen, E., Van Kempen, L., & Sahli, H. (2000). "Pattern Classiffcation Based on a Piecewise Multi-Linear Model for the Class Probability Densities. In SSPR 2000, SPR 2000, Proc. Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recog- nition and Statistical Pattern Recognition; Alicante, Spain; August 30 - September 1, 2000. (pp. 501-510). The Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recognition (SSPR 2000) and Statistical Pattern Recognition (SPR 2000), pp. 501-510, Alicante, Spain..
@inproceedings{b42065049cf9439094201e2605dd0565,
title = "{"}Pattern Classiffcation Based on a Piecewise Multi-Linear Model for the Class Probability Densities",
abstract = "When a Bayesian classifier is designed, a model for the class probability density functions (PDFs) has to be chosen. This choice is determined by a trade-off between robustness and low complexity -- which is usually satisfied by simple parametric models, based on a restricted number of parameters -- and the model's ability to fit a large class of PDFs -- which usually requires a high number of model parameters. In this paper, a model is introduced, where the class PDFs are approximated as piecewise multi-linear functions (a generalisation of bilinear functions for an arbitrary dimensionality). This model is compared with classical parametric and non-parametric models, from a point of view of versatility, robustness and complexity. The results of classification and PDF estimation experiments are discussed.",
author = "Edgard Nyssen and {Van Kempen}, Luc and Hichem Sahli",
note = "The Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recognition (SSPR 2000) and Statistical Pattern Recognition (SPR 2000), pp. 501-510, Alicante, Spain.; Unknown ; Conference date: 01-01-2000",
year = "2000",
month = aug,
day = "30",
language = "English",
pages = "501--510",
booktitle = "SSPR 2000, SPR 2000, Proc. Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recog- nition and Statistical Pattern Recognition; Alicante, Spain; August 30 - September 1, 2000.",
publisher = "The Joint IAPR Intl. Workshops on Syntactical and Structural Pattern Recognition (SSPR 2000) and Statistical Pattern Recognition (SPR 2000), pp. 501-510, Alicante, Spain.",
}