Crop and Fertilizer Recommendation System Applying Machine Learning Classifiers

Abstract

In today's agriculture, increasing crop yields while minimizing resource use is critical for sustainable and profitable agriculture. The program addresses the urgent need for personalized crop and fertilizer recommendations in various agricultural regions of India. Using machine learning techniques such as linear regression, random forests, k-nearest neighbors and outlier detection, our model aims to provide farmers with clear guidance based on various environmental factors and soil. Linear regression algorithm is used to establish the relationship between input and product yield, K-Neighbors adds recommendations by identifying the similarity between farm features and neighboring farm features. This holistic approach is designed to increase the credibility of our proposal, given the unique challenges posed by India's diversity and climate. The model accepts soil characteristics, climate, crop history and fertilizer use patterns. We worked to create a variety of useful and flexible products that can be more efficient and practical and provide good results to farmers in different regions. The need of the project is an easy-to-use website that provides farmers with information about crop selection and planting. By leveraging the power of machine learning, the program helps modern agriculture, promotes sustainable practices and increases the overall profitability of Indian agriculture.