EUVIMEDExplore EUVIMED
Open access · CC BY-NC-SA 4.0

Interpretable XAI-Driven Crop Price Prediction System Integrating Climatic and Market Dynamics for Trustworthy and Informed Agricultural Decision-Making

Mahesh Kumar R, Sukanya N S, Meenachi L, Vilas Namdeo Nitnaware, Rohini V, Kirubanand V. B

Journal of Applied Science and Technology Trends · 2025-12-31 · 222-235

DOI: 10.38094/jastt605653 ↗

At a glance

Journal article in Journal of Applied Science and Technology Trends, published online on 2025-12-31. Written by Mahesh Kumar R and 5 co-authors. The full text is freely available (open access, CC BY-NC-SA 4.0).

AbstractBibliographic detailsCiteRelated studies

Abstract (original language)

Stability of agriculture, food security and economic planning depend on precise price prediction of crops. By incorporating both climatic and market dynamics and explainable artificial intelligence (XAI), one can make agricultural systems more transparent and build trust in their predictions. Nevertheless, most current models of crop price forecasting are plagued by low interpretability, weak nonlinear dependence treatment, and they cannot provide information on how climatic and economic variables affect the results of the predictions. To overcome these shortcomings, the proposed Explainable Hybrid Deep Learning using SHAP-based Interpretability (EHDL-SHAP) framework uses Recurrent Neural Networks (RNNs) to find temporal climatic effects, and Gradient Boosted Regression Trees (GBRT) to identify non-temporal market dynamics. SHAP (SHapley Additive exPlanations) is also used to give explainable information by measuring the impact that each feature makes to the eventual prediction. The suggested system allows making agricultural choices based on the data and being transparent at the same time, correlating the predictions of the models with the reality of climatic and economic conditions. The experimental analysis of multi-regional crop data data shows that E HDL-SHAP has a stronger prediction accuracy, interpretability, and stakeholder confidence than traditional black-box models do. The framework is a smart, articulable and dependable instrument that farmers, policymakers, and agribusiness can use as a reference to make a substantial decision on crop management and pricing. The suggested approach significantly enhances the prediction performance of 94.8 percent, regional generalization performance by 0.9 percent or so, and the prediction performance of the temporal dependency captures the highest correlation.

Cite

APA

R, M. K., S, S. N., L, M., Nitnaware, V. N., V, R., & B, K. V. (2025). Interpretable XAI-Driven Crop Price Prediction System Integrating Climatic and Market Dynamics for Trustworthy and Informed Agricultural Decision-Making. Journal of Applied Science and Technology Trends, 222-235. https://doi.org/10.38094/jastt605653

Vancouver

R MK, S SN, L M, Nitnaware VN, V R, B KV. Interpretable XAI-Driven Crop Price Prediction System Integrating Climatic and Market Dynamics for Trustworthy and Informed Agricultural Decision-Making. Journal of Applied Science and Technology Trends. 2025:222-235. doi:10.38094/jastt605653

BibTeX
@article{R2025,
  author = {R, Mahesh Kumar and S, Sukanya N and L, Meenachi and Nitnaware, Vilas Namdeo and V, Rohini and B, Kirubanand V.},
  title = {Interpretable XAI-Driven Crop Price Prediction System Integrating Climatic and Market Dynamics for Trustworthy and Informed Agricultural Decision-Making},
  journal = {Journal of Applied Science and Technology Trends},
  year = {2025},
  pages = {222--235},
  doi = {10.38094/jastt605653}
}
Start for free →