IDTC: Crop Recommendation System using Improved Decision Tree Classification Algorithm

Crop prediction is essential for sustainable agriculture, and machine learning techniques have shown great assure in this area. In this paper, it presents an improved Decision Tree (DT) classification approach for crop prediction based on agricultural environmental factors. The proposed method leverages an original enhanced recursive feature selection method to identify the mainly relevant environmental parameter, such as soil type, temperature, humidity, with rainfall, which impact crop yields. The improved DT algorithm is then useful to categorize crops into different yield category. Experimental results demonstrate that the proposed approach outperforms conventional decision tree algorithm, achieving a high accuracy rate. The study also explores the impact of different feature selection techniques and DT parameter tuning on crop prediction accuracy. It has shown potential results with the model. The result found that the DT model performs well with a performance measure called MAE, R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>, Recall, Precision and F1 Score. The result shows that the forecast result of the DT representation is concurrent to expected information, which prove the efficiency of the model. So, this proposed method has significant implication for farmers, agricultural stakeholders, and policymakers, enabling data-driven decision making and optimized resource provision to improve crop yields.