• DocumentCode
    2951681
  • Title

    Development of Artificial Neural Network Models for Long-Range Meteorological Parameters Pattern Recognition over the Smaller Scale Geographical Region-District

  • Author

    Karmakar, S. ; Kowar, M.K. ; Guhathakurta, P.

  • Author_Institution
    Dept. of Comput. Applic., Bhilai Inst. of Technol., Durg
  • fYear
    2008
  • fDate
    8-10 Dec. 2008
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    Attempt to recognize pattern of meteorological parameters over the smaller scale geographical region (district) artificial neural network models have been developed. 54 years data for 1951-2004 have been used, of which the first 41 years (1951-1991) of data are used for training the network and data for the period 1991-2004 are used independently for validation. We have found that the mean absolute deviation (% of mean) between actual and predicted values of the each model is less than and half of the standard deviation (% of mean) in the independent period (1991-2004). The performances of these models in pattern recognition and prediction have been found to be extremely good. The models are developed and their evaluations have been presented in this paper.
  • Keywords
    geographic information systems; geography; geophysical techniques; meteorology; neural nets; pattern recognition; statistical analysis; artificial neural network model; long-range meteorological parameters pattern recognition; mean absolute deviation; smaller scale geographical region-district; Artificial neural networks; Chaos; Information systems; Meteorology; Neural networks; Paper technology; Pattern recognition; Predictive models; Region 10; Weather forecasting; Perception; artificial neural network; back-propagation; neuron; numerical modeling;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Industrial and Information Systems, 2008. ICIIS 2008. IEEE Region 10 and the Third international Conference on
  • Conference_Location
    Kharagpur
  • Print_ISBN
    978-1-4244-2806-9
  • Electronic_ISBN
    978-1-4244-2806-9
  • Type

    conf

  • DOI
    10.1109/ICIINFS.2008.4798370
  • Filename
    4798370