eduzhai > Applied Sciences > Food Science >

Application of Artificial Neural Networks in Prediction of Yielding of Rice in Bihar

  • Save

... pages left unread,continue reading

Document pages: 4 pages

Abstract: Rice crop have highest share approx 40 in overall food grains in India and 46 in Bihar and Its provide food security to the region. The crop yield is reliant on favourable climate. Due to fast climate changing and growing world population crop yield prediction play a vital role in decision making and planning of food in future. In different climate conditions the prediction of crop productivity, can lend a hand to farmers and stakeholders in production decisions for agronomy and crop selection.ANN is used in this study to forecast rice production yielding as well as look into the factor by which the rice yield is affected for various areas of Bihar, India. All the statistics are collected from openly accessible records of Government of India and Govt of Bihar of 38 Districts of Bihar, India. The study parameters are Max, Mean, Min temperature; rainfall, rice crop evaporation and transpiration, production, vicinity and yield in Kharif season (July to December) for the years (2012-2016). WEKA tool is used for dataset processing. Multilayer Perceptron Neural Network developed and to validate statistics Cross validation process was used. The result shows that the accurateness of 94.5 with a sensitivity of 93.3 and specificity of 97.3 . For study after that relative absolute, mean absolute, Root mean Square and, root relative square error to be calculated. The act of classifier is shown pictorially briefed using ROC curve.ROC Curve obtained from the designed module is compared with other module of ANN and it offers high yield of Rice produced during Kharif season.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×