Hybrid Uncertainty Aware Model for Precision Crop Recommendation System Using Machine Learning

DOI

10.2991/978-94-6239-713-2_25How to use a DOI?

Keywords

Crop Recommendation; Ensemble Learning; Explainable AI; Precision Agriculture; Tabular Machine Learning; Uncertainty Quantification

Abstract

The presented paper proposes a hardware-independent, uncertainty-aware crop recommendation system that integrates agronomic, climatic, and economic factors to produce precise, profit-oriented recommendations. The system takes extensive preprocessing and feature engineering to analyze the nutrient profiles in the soil, weather conditions, past yields, and market indicators. The hybrid stacked ensemble model that integrates Gradient-Boosted Decision Trees with LightGBM, XGBoost, CatBoost, and a Deep Tabular Network (TabNet) is adopted to uncover the complex interactions among features in the model and to provide interpretability. The quantification of uncertainty is implemented by Monte Carlo Dropout and NGBoost, which are incorporated in a cost-sensitive loss that minimizes risky recommendations. SHAP-based explanations further improve model transparency by providing insight into feature contributions. The model is tested on the Kaggle Crop Recommendation dataset, comprising 2,200 samples of 22 crops. In this paper, I propose a hybrid stacked ensemble comprising XGBoost, LightGBM, CatBoost, and TabNet, with Monte Carlo Dropout, to recommend crops based on uncertainty, achieving 99.39% accuracy. The following are the novel features: (1) hybrid GBDT and deep tabular learning, (2) cost-sensitive uncertainty quantification, and (3) SHAP-based explainability. The results show that ensemble learning can be used with deep tabular models and uncertainty-aware optimization to offer robust and reliable agricultural decision support systems.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Atul Kumar
AU  - Jitendra Kumar
AU  - Nikhil Pratap Singh
AU  - Jayant Kumar
AU  - Sanchita Adhikari
AU  - Om Prakash Yadav
PY  - 2026
DA  - 2026/06/25
TI  - Hybrid Uncertainty Aware Model for Precision Crop Recommendation System Using Machine Learning
BT  - Proceedings of the International Conference on Advances in Computing Technology and Artificial Intelligence (COMPUTATIA 2026)
PB  - Atlantis Press
SP  - 336
EP  - 350
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-713-2_25
DO  - 10.2991/978-94-6239-713-2_25
ID  - Kumar2026
ER  -