A Generalized Extreme Value Naïve Bayes Framework for Classifying Univariate Extreme Value
DOI:
https://doi.org/10.37375/bsj.v7i20.3657Keywords:
Extreme value theory, Generalized Extreme Value distribution, Naïve Baayes, Machine LearningAbstract
This paper aims to develop a model that predicts the class of extreme values by utilizing the Naïve Bayes algorithm and employing statistical models from Extreme Value Theory (EVT). It aligns with recent interests in creating innovative algorithms that integrate both machine learning (ML) and EVT. The model is developed using simulated data, and the experimental results demonstrate significant performance, achieving an accuracy, sensitivity, and specificity of 0.98, along with an AUC of 0.97. The MATLAB software package is used to model and implement the proposed algorithm, which establishes a classification rule for determining the class of extreme values.
References
Alpaydın, E. (2010). Introduction to machine learning (2nd). Massachusetts: MIT Press.
Castillo-Mateo, J., Asín, J., Cebrián, A. C., Mateo-Lázaro, J., & Abaurrea, J. (2023). Bayesian variable selection in generalized extreme value regression: Modeling annual maximum temperature. MDPI.
Coles, S. (2001). An introduction to statistical modeling of extreme values (3rd ed.). London: Springer.
Duda, R. O., Hart, P. E., & Stork, D. G. (2001). Pattern classification (2nd ed.). New York: John Wiley & Sons.
Emtair, M. (2015). On a supervised classifier that can learn better from mixed peaks-over-threshold. International Journal of Research Studies.
Emtair, M. (2016). Numerical solution to the problem of optimal boundary of generalized Pareto mixtures. International Journal of Applied Science and Engineering Technology.
Ferreira, A., & de Haan, L. (2006). Extreme Value Theory: An Introduction. New York: Springer.
Ferreira, A., & de Haan, L. (2014). On the block maxima method in extreme value theory: PMM estimators. The Annals of Statistics.
Huet, N., & Prosdocimi, I. (2025). Robust and efficient estimation for the generalized extreme-value distribution with application to flood frequency analysis in the UK.
Ktakauer, N. (2024). Extended blended generalized extreme value distribution. Journal of Statistical Distributions and Applications, 12(3), 123-145.
Murphy, K. (2012). Machine Learning: A Probabilistic Perspective. Massachusetts: MIT Press.
Martin, E., & Stedinger, J. (2000). Generalized maximum-likelihood generalized extreme-value quantile estimators for hydrologic data. Water Resources Research.
Sabourin, A. (2021). Extreme value theory and machine learning: A year of research at Telecom Paris. Telecom Paris.
Reiss, R., & Thomas, M. (2007). Statistical Analysis of Extreme Values. Switzerland: Birkhäuser Verlag.
