Clustering-Based Multi-Crop Recommendation System for Precision Agriculture

Agriculture in India is an ancient and essential industry, providing food and livelihoods to millions. However, centuries of continuous land use without sufficient knowledge of soil properties have depleted soil nutrients, leading to declining crop yields. Limited adoption of modern agricultural practices further intensifies this issue, underscoring the need for precision agriculture. To address this, we analyzed a dataset containing soil nutrient content and key climatic factors such as temperature, humidity, and rainfall, enabling optimal crop recommendations to enhance farmers’ yields. Using cluster analysis, this paper identified similarities in crop requirements, promoting multi-cropping on the same land to boost productivity, especially for small-scale farmers. Among the models tested, the Random Forest and Naive Bayes algorithms showed remarkable effectiveness, achieving an accuracy of 99%.