An EWOA-WTDCNN Model: Effective Whale Optimization Algorithm Incorporating with Weight-Tuned Deep Convolutional Neural Network for Forecasting Indian Crop Yielding Process

Abstract

Field-level crop yield prediction (CYP) is essential for quantitative and economic analysis when developing agricultural commodity plans for import-export strategies and raising farmer incomes. Crop breeding has traditionally been a time- and money-intensive process. To predict increased agricultural output, CYP was developed. This research suggests effective dimensionality reduction (DR) and deep learning (DL) techniques for CYP for Indian regional crops. The three phases of this paper were preprocessing DR and classification. The dataset is first used to gather information about South Indian agriculture. Data cleaning and normalization are then applied as preprocessing to the gathered dataset. Squashed exponential kernel-based principal component analysis (SEKPCA) is then used to perform the DR. CYP, which forecasts the high crop yield profit, is built on a weight-tuned deep convolutional neural network (WTDCNN) with an Effective Whale Optimisation Algorithm (EWOA). According to the simulation results, the suggested strategy achieves greater performance for CYP compared to existing methods with enhanced accuracy.