Title: An efficient crop yield prediction using Bootstrap Aggregative MapReduce Rocchio Classification
Authors: C.P. Saranya; Sini Raj Pulari; Shriram K. Vasudevan
Addresses: Department of Computer Science and Engineering, Coimbatore Institute of Engineering and Technology, Coimbatore 641109, India ' Bahrain Polytechnic, Isa Town, Bahrain ' K. Ramakrishnan College of Technology, Kariyamanikam Road, Samayapuram, Trichy – 621112, India
Abstract: Climatic changes in environmental condition affect crop growth and crop yield. Therefore, the impact of climate change on crop production is a significant problem to be solved. Though a handful of research works are present to predict the crop yield, the prediction accuracy of crop productivity is not sufficient. Besides, the time duration needed to identify the crop production using big agriculture dataset was higher. To avoid such limitations, the Bootstrap Aggregative MapReduce Rocchio Classification (BAMRC) technique is proposed. The BAMRC technique contains two key processes i.e., feature selection and classification. The experimental evaluation of BAMRC technique is conducted on metrics such as prediction accuracy, prediction time and false positive rate relating to numerous test instances collected at varied time period. The experimental results depict that the BAMRC technique is able to improve the prediction accuracy and also minimise the prediction time of crop yields from soybean large dataset when compared to state-of-the-art works.
Keywords: bagging; big agriculture data; climate; correlation; crop yield; features; Hadoop MapReduce function; strong classifier; AI; artificial intelligence for agriculture.
International Journal of Sustainable Agricultural Management and Informatics, 2021 Vol.7 No.2, pp.129 - 148
Received: 16 Jun 2020
Accepted: 19 Sep 2020
Published online: 30 Jun 2021 *