Abstract
Agriculture plays a vital role in the economy, but farmers often face challenges like unpredictable weather, poor soil conditions, and lack of proper guidance. These issues can affect crop production and lead to losses. To address this, this paper presents a smart agriculture system using Artificial Intelligence (AI) to help farmers make better decisions. Farmers face challenges such as climate variability, soil degradation, and lack of decision support. This paper proposes an AI-based smart agriculture system for crop recommendation, yield prediction, and rainfall forecasting. Machine learning models including Random Forest, XG Boost, and Artificial Neural Networks are trained on agricultural data (temperature, rainfall, humidity, and soil nutrients). The system achieves up to 98% accuracy for crop recommendation and 96% for rainfall prediction. A key contribution is the integration of Explainable AI (XAI) using SHAP to interpret predictions and increase user trust. The proposed system supports better decision-making, improves productivity, and promotes sustainable farming.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by-nc. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 2
Claim · model-written summary
The proposed system achieves up to 98% accuracy for crop recommendation.
Verbatim
e0“The system achieves up to 98% accuracy for crop recommendation”
Claim · model-written summary
The proposed system achieves 96% accuracy for rainfall prediction.
Verbatim
e0“The system achieves up to 98% accuracy for crop recommendation and 96% for rainfall prediction”