Abstract
In this paper, we explore the problem of estimating lower and upper densities from imprecisely defined families of parametric kernels. Such estimations allow to rely on a single bandwidth value, and we show that it provides good results on classification tasks when extending the naive Bayesian classifier.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
A kernel is here a symmetric, non-negative function with \(\int _{\mathbb {R}} K(y)dy=1\) and mean 0.
- 2.
In some sense, to regularize our model.
References
Corani, G., Zaffalon, M.: Learning reliable classifiers from small or incomplete data sets: the naive credal classifier 2. J. Mach. Learn. Res. 9, 581–621 (2008)
Destercke, S., Strauss, O.: Filtering with clouds. Soft Comput. 16(5), 821–831 (2012)
Silverman, B.W.: Density estimation for statistics and data analysis, vol. 26. CRC Press, Boca Raton (1986)
Wand, M.P., Jones, M.C.: Kernel smoothing. CRC Press, Boca Raton (1994)
Zaffalon, M.: The naive credal classifier. J. Probab. Plann. Infer. 105, 105–122 (2002)
Zaffalon, M., Corani, G., Mauá, D.: Evaluating credal classifiers by utility-discounted predictive accuracy. Int. J. Approx. Reason. 53(8), 1282–1301 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Dendievel, G., Destercke, S., Wachalski, P. (2019). Density Estimation with Imprecise Kernels: Application to Classification. In: Destercke, S., Denoeux, T., Gil, M., Grzegorzewski, P., Hryniewicz, O. (eds) Uncertainty Modelling in Data Science. SMPS 2018. Advances in Intelligent Systems and Computing, vol 832. Springer, Cham. https://doi.org/10.1007/978-3-319-97547-4_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-97547-4_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-97546-7
Online ISBN: 978-3-319-97547-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)