Smart Agriculture Using Machine Learning

Abstract

Agriculture is a vital source of income in India, but
farmers often face challenges in selecting the right
crop and fertilizer, leading to reduced productivity.
Smart agriculture helps address this by leveraging
soil data, crop yield statistics, and environmental
factors to provide accurate recommendations. This
project proposes a machine learning-based
recommendation system using majority voting
(Random Forest, Naive Bayes, SVM, and Logistic
Regression) to suggest the most suitable crop
based on site-specific parameters. Additionally, a
fertilizer recommendation system compares userinputted
soil nutrient levels with optimal values,
identifying deficiencies and providing targeted
suggestions to enhance soil fertility. The project
also integrates a disease prediction model that
detects potential crop diseases early, enabling
timely intervention and reducing crop loss. By
combining crop selection, fertilizer
recommendation, and disease prediction, this
system empowers farmers with data-driven
insights, ultimately improving agricultural
productivity, resource efficiency, and
sustainability.