Ensemble Machine Learning Model for Better Crop Production

The world’s food supply is primarily dependent on agriculture. In addition, it provides the raw materials needed by other businesses. The growth in agriculture cannot cope with the world’s population which is increasing day by day. Besides increasing food production for a developing country that has limited land and resources, can cause a shortage of food in the near future. Selecting the right crop for a specific region is crucial for enhancing its production. Historical data is used to accurately anticipate the area’s agricultural yield based on measures of soil components (nitrogen, phosphorus, and potassium), and climatic measures (temperature, humidity, rainfall, and pH). We then suggest using an ensemble machine learning technique known as ‘Naive Forest’ to accurately forecast the harvests. After investigating five classical machine learning algorithms and ensemble learning (Random Forest) algorithms and their respective accuracy, this ‘Naive Forest’ approach is designed. We think that the suggested model will support agricultural workers and farmers inappropriately producing crops.