Modelling Rainfall Prediction using Bayesian Approach with Inter-Station Dependencies
DOI:
https://doi.org/10.18486/ijcsnt/13.2.176Keywords:
Bayesian Network, Bayesian Classification, Rainfall Forecast, Rainfall PredictionAbstract
Rainfall forecasting has been one of the most challenging problem for scientists and engineers around the globe. A rainfall forecast is highly beneficial for efficient management of surface and subsurface water resources in arid regions. Rainfall is a complex hydrological phenomenon which depends on many atmospheric processes. These atmospheric conditions are generally used for rainfall prediction. Two atmospheric variables, Temperature and Cloud Cover, are used in this study for rainfall prediction. Rainfall distribution is not uniform among different stations but stations share a strong correlation, and this property has been used for improve the prediction accuracy of a model. Bayesian network offers framework to incorporate the conditional dependencies between the nodes/variables. K2 algorithm is used to finding out the rainfall dependencies among the stations. Thirty-two stations of Rajasthan, India are used for this study. Rainfall, temperature and cloud cover data of 102 years has been used for this study which is divided into standard 70:30 ratio for training and testing respectively. The prediction results of both cases (considering inter-station rainfall dependencies and without considering inter-station rainfall dependencies) are compared and an improvement in accuracy was found at most of the stations while considering inter-station dependencies.
References
Ben-Gal I. (2007) Bayesian Networks in Encyclopedia of Statistics in Quality & Reliability. Wiley & Sons. DOI: https://doi.org/10.1002/9780470061572.eqr089
Heckerman David (1997) Bayesian Networks for Data Mining. Data Mining and Knowledge Discovery 1, 79–119 (1997). DOI: https://doi.org/10.1023/A:1009730122752
Cooper Gregory E., Herskovits Edwards (1992) A Bayesian Method for the Induction of Probabilistic Networks from Data. Machine Learning, 9, 309–347. DOI: https://doi.org/10.1023/A:1022649401552
Cheng Jie, Bell David A., Liu Weiru (1997) An Algorithm for Bayesian Belief Network Construction from Data. In the Proceedings of AI & STAT’97.
Borsuk Mark E., Stow Craig A., Reckhow Kenneth H. (2004) A Bayesian Network of Eutrophication Models for Synthesis, Prediction, and Uncertainty Analysis. Ecological Modelling, 173 (2004), 219–239. DOI: https://doi.org/10.1016/j.ecolmodel.2003.08.020
Cofino Antonio S., Cano Rafael, Sordo Carmen, Gutierrez Jose M. (2002) Bayesian Networks for Probabilistic Weather Prediction. In Proceedings of the 15th European Conference on Artificial Intelligence (ECAI 2002), IOS Press, pp. 695–700.
Cano Rafael, Sordo Carmen, Gutierrez Jose M. (2004) Applications of Bayesian Networks in Meteorology. In Advances in Bayesian Networks, Gámez et al. (Eds.), Springer, pp. 309–327. DOI: https://doi.org/10.1007/978-3-540-39879-0_17
Nikam Valmik B., Meshram B.B. (2013) Modeling Rainfall Prediction Using Data Mining Method. In 2013 Fifth International Conference on Computational Intelligence, Modelling and Simulation. DOI: https://doi.org/10.1109/CIMSim.2013.29
Nandar, A. (2009, December). Bayesian Network Probability Model for Weather Prediction. In Current Trends in Information Technology (CTIT), 2009 International Conference on, pp. 1–5. IEEE. DOI: https://doi.org/10.1109/CTIT.2009.5423132
Sharma Ashutosh, Goyal Manish Kumar (2015) Bayesian Network Model for Monthly Rainfall Forecast. In Proceedings of the 2015 IEEE International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN), Kolkata, India, November 2015. DOI: https://doi.org/10.1109/ICRCICN.2015.7434243
Rathore, M. S. (2004). State Level Analysis of Drought Policies and Impacts in Rajasthan, India (Vol. 93). IWMI.
Murphy Kevin P. (2001) The Bayes Net Toolbox for Matlab. Computing Science and Statistics, Vol. 33, 2001. Code available at https://code.google.com/p/bnt/
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Ashutosh Sharma, Manish Kumar Goyal
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
