A ML-AI ENABLED ENSEMBLE MODEL FOR PREDICTING AGRICULTURAL YIELD

• Published in: Cogent food & agriculture Vol. 8; no. 1
• Main Authors: Kundu, Seeboli Ghosh, Ghosh, Anupam, Kundu, Avisek, G P, Girish
• Format: Journal Article
• Language: English
• Published: London Cogent 31-12-2022; Taylor & Francis Ltd; Taylor & Francis Group
• Subjects: Agricultural production; Agriculture; Algorithms; Artificial intelligence; Crop yield; Crops; ensemble method; forecasting agricultural yield; Irrigation; Machine learning; ML models; Predictions; Rainfall; Soils; Support vector machines
• ISSN: 2331-1932; 2331-1932
• DOI: 10.1080/23311932.2022.2085717

Simplistic linear methods for predicting crop yield leave out important factors like climate, rainfall, soil, irrigation, and land characteristics. Recent literature points to use of individual Machine Learning (ML) and Artificial Intelligence (AI) models for better prediction of crop yield. However, such methods have not been used in the Indian context. Moreover, given the diversity of land, climate, soil and irrigation facilities in the country, it is necessary to develop an ensemble approach incorporating a significant number of ML algorithms to have a better prediction of crop yield across geographies of the country. The purpose of this paper is to: (a) develop scenario-specific algorithms for ML models and identify the best fit model for yield prediction and (b) develop an ensemble approach synthesizing the ML models for better overall prediction of crop yield in India.