Gorilla troops optimization with deep learning based crop recommendation and yield prediction

Abstract

Agriculture plays a vital role in the Indian economy. Crop recommendation for a specific region is a tedious process as it can be affected by various variables such as soil type and climatic parameters. At the same time, crop yield prediction was based on several features like area, irrigation type, temperature, etc. The latest breakthroughs in Machine Learning (ML) and Artificial Intelligence (AI) technologies pave the way to designing effective crop recommendation and prediction models. Despite the significant advancements of Deep Learning (DL) models in crop recommendation, hyperparameter tuning using metaheuristic algorithms becomes essential for enhanced performance. This tool allows users to anticipate appropriate crops and their expected yields for a provided year, assisting agriculturalists in choosing crops suitable for their area and period and anticipating productivity. This article introduces a Gorilla Troops Optimization with Deep Learning-based Crop Recommendation and Yield Prediction model (GTODL-CRYPM). The proposed GTODL-CRYPM model mainly focuses on two processes, namely, crop recommendation and crop prediction. Firstly, the GTO with Long Short-Term Memory (LSTM) technique is employed to make efficient crop recommendations. Besides, the GTO model is applied to adjust the LSTM parameters optimally. Next, the Deep Belief Network (DBN) technique was executed to predict crop yield accurately. A wide range of experiments have been conducted to report the improved performance of the GTODL-CRYPM model. The outcomes are examined under the Crop Recommendation Dataset and Crop Yield Prediction Dataset. Experimentation outcomes highlighted the significant performance of the GTODL-CRYPM approach on the compared approaches, with a maximum accuracy of 99.88% and an R2 score of 99.14%.

Keywords

• Agriculture
• Crop yield prediction
• Crop recommendation
• Deep learning
• Gorilla troops optimizer