Smart Crop Recommendation Using Ensemble Machine Learning Models

Abstract

Agriculture is the primary occupation for a huge portion of Pakistan's population and plays a vital role in the country’s economic growth and food security. Crop growth is significantly influenced by various environmental and soil-related factors such as weather, chemical inputs, soil moisture, phosphorus levels, humidity, temperature, and rainfall. To enhance crop productivity and decision-making, this research proposes a smart crop recommendation system based on ensemble learning using supervised machine learning models. Sensor data is used to monitor key factors, which are then analyzed using an ensemble of different supervised learning techniques. By combining the strengths of multiple models through a voting-based approach, more accurate recommendations are generated. Among the evaluated models, decision trees and artificial neural networks provided the most effective results, with the artificial neural network achieving an accuracy of up to 98%. This approach supports the development of precision agriculture, which emphasizes site-specific crop management using modern agricultural technologies. Precision farming is gradually gaining traction in developing countries like Pakistan, offering improved efficiency and sustainability in agriculture.