Abstract
For recent several years, we have suffered from a variety of unusual weather phenomena. In particular, regional torrential rains have caused an immeasurable losses of life and property, and the forecast of heavy rainfall becomes progressively important as time goes on. We study wrapper-based genetic feature selection using machine learning techniques such as SVM or k-NN for very short-term heavy rainfall prediction in the southern part of the Korean Peninsula. Historical weather data were collected from 408 AWSes of the Korea Meteorological Administration during recent 4 years. The data from 2007 to 2008 were selected to train SVM and k-NN models and the data of the year 2009 were used as a validation set in our genetic algorithm, and the data of the year 2010 were used as a test set. Undersampling is to match the number of samples of a high frequency with that of a low frequency. We undersampled the train set into two classes: heavy rainfall (more than 70mm for 6 hours or more than 110mm for 12 hours) and the other. Test without undersampling produced low ETS and took too long time. The validation set was used to choose the important ones among 72 features using a genetic algorithm. Normalized data between 0 and 1 had a good influence on the performance compared to the test without normalization, and especially on SVM. The SVM using important features performed about 3.5 times better than that using all features.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Korea Meteorological Administration, http://www.kma.go.kr
Seo, J.H., Kim, Y.H.: A survey on rainfall forecast algorithms based on machine learning technique. In: Proceedings of KIIS Fall Conference, vol. 21(2), pp. 218–221 (2011) (in Korean)
French, M.N., Krajewski, W.F., Cuykendall, R.R.: Rainfall forecasting in space and time using a neural network. Journal of Hydrology 137(1-4), 1–31 (1992)
Toth, E., Brath, A., Montanari, A.: Comparison of short-term rainfall prediction models for real-time flood forecasting. Journal of Hydrology 29, 132–147 (2000)
Burian, S.J., Durrans, R., Nix, S.J., Pitt, R.E.: Training artificial neural networks to perform rainfall disaggregation. Journal of Hydrologic Engineering 6(1), 43–51 (2001)
Ramirez, M.C.V., Velho, H.F.C., Ferreira, N.J.: Artificial neural network technique for rainfall forecasting applied to the Sao Paulo region. Journal of Hydrology 301, 146–162 (2005)
Hung, N.Q., Babel, M.S., Weesakul, S., Tripathi, N.K.: An artificial neural network model for rainfall forecasting in Bangkok, Thailand. Hydrology and Earth System Science 5, 183–218 (2008)
Ingsrisawang, L., Ingsriswang, S., Somchit, S., Aungsuratana, P., Khantiyanan, W.: Machine learning techniques for short-term rain forecasting system in the northeastern part of Thailand. Proceedings of World Academy of Science, Engineering and Technology 31, 248–253 (2008)
Hong, W.C.: Rainfall forecasting by technological machine learning models. Journal of Applied Mathematics and Computing 200(1), 41–57 (2008)
Kishtawal, C.M., Basu, S., Patadia, F., Thapliyal, P.K.: Forecasting summer rainfall over India using genetic algorithm. Geophysical Research Letters 30(23), 1–5 (2003)
Liu, J.N.K., Li, B.N.L., Dillon, T.S.: An improved naïve Bayesian classifier technique coupled with a novel input solution method. IEEE Transactions on Systems, Man, and Cybernetics−Part C: Applications and Reviews 31(2), 249–256 (2001)
Chang, R., Pei, Z., Zhang, C.: A modified editing k-nearest neighbor rule. Journal of Computers 6(7), 1493–1500 (2011)
Yin, Y., Han, D., Cai, Z.: Explore data classification algorithm based on SVM and PSO for education decision. Journal of Convergence Information Technology 6(10), 122–128 (2011)
Choi, Y.S., Moon, B.R.: Feature selection in genetic fuzzy discretization for the pattern classification problems. IEICE Transactions on Information and Systems 90(7), 1047–1054 (2007)
De Jong, K.A.: An analysis of the behavior of a class of genetic adaptive systems, PhD thesis, University of Michigan (1975)
Goldberg, D.E.: Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley Professional (1989)
Yoon, G.M., Kim, J., Kim, Y.H., Moon, B.R.: Performance improvement by genetic feature selection and adjusting ratings’ mid-point value in the neural network-based recommendation models. Advances in Information Sciences and Service Sciences (March 2012) (in Press)
Automatic Weather Stations, http://www.automaticweatherstation.com
Chawla, N.V.: Data mining for imbalanced datasets: an overview. In: Data Mining and Knowledge Discovery Handbook, vol. 5, pp. 853–867 (2006)
Chang, C., Lin, C.: LIBSVM: a library for support vector machines. ACM Transactions on Intelligent Systems and Technology 2, 1–27 (2011)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Seo, JH., Kim, YH. (2012). Genetic Feature Selection for Very Short-Term Heavy Rainfall Prediction. In: Lee, G., Howard, D., Kang, J.J., Ślęzak, D. (eds) Convergence and Hybrid Information Technology. ICHIT 2012. Lecture Notes in Computer Science, vol 7425. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32645-5_40
Download citation
DOI: https://doi.org/10.1007/978-3-642-32645-5_40
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32644-8
Online ISBN: 978-3-642-32645-5
eBook Packages: Computer ScienceComputer Science (R0)