AN ENSEMBLE ALGORITHM FOR CROP YIELD PREDICTION

Computer Science and Engineering

Agriculture has a significant part in the Indian economy and jobs. The most common problem faced by Indian farmers is that they do not select crops based on soil requirements, resulting in major productivity concerns. Machine learning algorithms will assist farmers in determining which crop to plant in order to provide the most yield by taking into account parameters such as temperature, rainfall, location, and so on. Precision agriculture can help overcome this challenge. This method considers three elements: soil characteristics, soil kinds, and crop yield data gathering. Based on these parameters, the farmer is recommended a crop to produce. This system includes a model for estimating crop yield that is exact and accurate. This system includes a model that is exact and accurate in predicting crop production and providing the end-user with the appropriate fertilizer ratio recommendations based on atmospheric and soil factors of the land, hence increasing crop yield and farmer revenue. As a result, the suggested system's pre-processing methods are based on feature scaling techniques including Min-Max Scaling, Standardization, and Normalization, which help to extract data on soil quality and weather-related information as an input. The soil's quality, such as nitrogen, phosphorus, potassium, pH, Rainfall, temperature, and humidity are all weather-related variables that can be used to predict a better harvest. We are using datasets from the Kaggle platform for our project. Using the ensemble modeling approach, we were able to do this. Different machine learning techniques, such as the Random Forest algorithm, Adaboost Classifier, Gradient Boosting Classifier, and K-Nearest Neighbors, were compared.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.