.svg)
Current issue: Volume 24, No. 2 (2024)
DOI: https://doi.org/10.59893/abud.24(2).007
Jorvekar P.P., Wagh S.K., Prasad J.R. 2024. Crop yield prediction using deep learning techniques in the agriculture domain. Acta Biol. Univ. Daugavp., 2024(2): 223-237
Abstract
In agriculture, yield prediction is a critical issue as all farmers would like to know how much harvest they may expect. In past decades, yield predictions were made by considering the farmer’s previous profitability with that specific crop and field. The implementation of machine learning techniques can help with the prediction of yield, which is a significant challenge that remains to be solved using the information currently available. In agriculture, many machine-learning approaches are employed and assessed to forecast crop yield. An agricultural yield prediction system is proposed and developed in this work using historical data. This is achieved by using deep learning algorithms for agriculture data, such as Independent Component Analysis (ICA) with Crow Search Optimization Algorithm (CSOA) and Deep Convolutional Neural Network (DCNN), and suggesting fertilizer optimal for each crop. The suggested study uses a DCNN classification method over the ICA-CSO approach to estimate agricultural production. The suggested approach outperforms existing models and predicts agricultural output with 97 percent accuracy while maintaining the baseline data distribution, giving an accurate perspective of forecasting crop yields using deep learning algorithms.
Keywords: Independent Component Analysis, Crow Search Optimization Algorithm.
*Corresponding author: Priti Prakash Jorvekar. Smt. Kashibai Navale College of Engineering, SPPU University, Pune, India. E-mail: pritiprmjorvekar@gmail.com
Sharmila Kishor Wagh. MES College of Engineering, Pune, SPPU University Pune, India
Jayashree Rajesh Prasad. School of Computing, MIT Art Design and Technology University Pune, India
