Call for Paper - January 2023 Edition
IJCA solicits original research papers for the January 2023 Edition. Last date of manuscript submission is December 20, 2022. Read More

Comparison of Nearest Neighbor (ibk), Regression by Discretization and Isotonic Regression Classification Algorithms for Precipitation Classes Prediction

Print
PDF
International Journal of Computer Applications
© 2014 by IJCA Journal
Volume 96 - Number 21
Year of Publication: 2014
Authors:
Solomon Mwanjele Mwagha
Masinde Muthoni
Peter Ochieng
10.5120/16919-6729

Solomon Mwanjele Mwagha, Masinde Muthoni and Peter Ochieng. Article: Comparison of Nearest Neighbor (ibk), Regression by Discretization and Isotonic Regression Classification Algorithms for Precipitation Classes Prediction. International Journal of Computer Applications 96(21):44-48, June 2014. Full text available. BibTeX

@article{key:article,
	author = {Solomon Mwanjele Mwagha and Masinde Muthoni and Peter Ochieng},
	title = {Article: Comparison of Nearest Neighbor (ibk), Regression by Discretization and Isotonic Regression Classification Algorithms for Precipitation Classes Prediction},
	journal = {International Journal of Computer Applications},
	year = {2014},
	volume = {96},
	number = {21},
	pages = {44-48},
	month = {June},
	note = {Full text available}
}

Abstract

Selection of classifier for use in prediction is a challenge. To select the best classifier comparisons can be made on various aspects of the classifiers. The key objective of this paper was to compare performance of nearest neighbor (ibk), regression by discretization and isotonic regression classifiers for predicting predefined precipitation classes over Voi, Kenya. We sought to train, test and evaluate the performance of nearest neighbor (ibk), regression by discretization and isotonic regression classification algorithms in predicting precipitation classes. A period of 1979 to 2008 daily Kenya Meteorological Department historical dataset on minimum/maximum temperatures and precipitations for Voi station was obtained. Knowledge discovery and data mining method was applied. A preprocessing module was designed to produce training and testing sets for use with classifiers. Isotonic Regression, K-nearest neighbours classifier, and RegressionByDiscretization classifiers were used for training training and testing of the data sets. The error of the predicted values, root relative squared error and the time taken to train/build each classifier model were computed. Each classifier predicted output classes 12 months in advance. Classifiers performances were compared in terms of error of the predicted values, root relative squared error and the time taken to train/build each classifier model. The predicted output classes were also compared to actual year classes. Classifier performances to actual precipitation classes were compared. The study revealed that the nearest neighbor classifier is a suitable for training rainfall data for precipitation classes prediction.

References

  • Ashok, M. et al, 2006. Linking Seasonal Climate Forecasts with Crop Simulation to Optimize Maize Management, CCSP Workshop: Climate Science in Support of Decision Making, 14-16 November 2005 Crystal Gateway Marriott Arlington, Virginia 14-16 November 2005
  • Dostrani, M. et al (2010). Application of ANN and ANFIS Models on Dryland Precipitation Prediction (Case Study: Yazd in Central Iran). Journal of Applied Sciences, 10: 2387-2394.
  • Gong, Z. et al, 2010. Risk Prediction of Agricultural Drought in China. 2010 Seventh International Conference on Fuzzy Systems and Knowledge Discovery (FSKD 2010). Kenya Meteorological Department, Agrometeorological bulletin, Issue No. 27/2009.
  • Kozyra, J. et al 2009. Institute Of Soil Science and Plant Cultivation National Research Institute, International Symposium, Climate change and Adaptation Options in Agriculture, Viena, June, 22-23 2009.
  • Ladislaus B. et al, 2010. Indigenous knowledge in seasonal rainfall prediction in Tanzania: A case of theSouth-western Highland of Tanzania, Journal of Geography and Regional Planning Vol. 3(4), pp. 66-72,April 2010.
  • Lin Zhu, Jing M. Chen, Qiming Qin, Mei Huang, Lianxi Wang, Jianping Li, Bao Cao: Assimilating Remote Sensing based Soil Moisture in an Ecosystem Model (BEPS) for Agricultural Drought Assessment. IGARSS (5) 2008: 437-440
  • Niu Shulian; Susaki Junichi, 2006. Detection of Agricultural Drought in Paddy Fields Using NDVI from MODIS Data. A Case Study in Burirum Province, Thailand.
  • Patrick O, 2006. Agricultural Policy in Kenya: Issues and Processes, A paper for the Future Agricultures Consortium workshop, Institute of Development Studies, 20-22 March 2006.
  • Peter Reutemann, (2007). WEKA Knowledge Flow Tutorial for Version 3-5-7, University of Waikato 2007
  • Tsegaye T. & Brian W. 2007. The Vegetation Outlook (VegOut): A New Tool for Providing Outlooks of General Vegetation Conditions Using Data Mining Techniques. ICDM Workshops 2007: 667-672
  • Z. (Bob) Su, Y. Chen, M. Menenti, J. Sobrino, Z. -L. Li, W. Verhoef, L. Wang, Y. Ma, L. Wan, Y. He, Q. H.
  • Liu, C. Li, J. WEN, R. van der Velde, M. van Helvoirt, W. Lin, X. Shan, 2007. Drought Monitoring and Prediction over China, In: Proceedings of the 2008 Dragon symposium, Dragon programme, final results, 2004-2007, Beijing, China 21-25 April 2008.