India's socio-economic well-being hinges significantly on agriculture, necessitating innovative solutions. While traditional farming has always utilized data, modern challenges demand more sophisticated approaches. Leveraging data mining, we construct machine learning models crucial for addressing agricultural complexities. India has now surpassed China to become the most populous country in terms of population. As we look towards the future, it is projected that the global population will keep growing. By the year 2050, the Earth will have an additional 9 billion people. This means that we urgently need to increase food production by 70%, while also using less land. India's expensive agricultural sector encounters hurdles stemming from inefficient fertilizer usage. Insufficient understanding of soil nutrients and environmental factors leads to wastage and subpar yields. Achieving optimal yields requires balancing fertilizer doses, a task influenced by diverse environmental and soil conditions. Therefore, a fertilizer prediction model is indispensable for maximizing crop productivity. This paper adopts a machine learning approach to predict optimal fertilizer varieties. Analyzing soil features (N, P, K, pH) and environmental parameters (temperature, humidity, rainfall), the model recommends suitable fertilizers for various crops. Utilizing techniques like K-Nearest Neighbors, Decision Trees, Random Forests, and Gradient Boosting, we aim to provide accurate predictions. This paper implies that the Random Forest classifier is the most suitable algorithm for predicting the suitable variety of fertilizer for the climatic and field conditions given by the farmers with an accuracy of 95%. By leveraging available datasets, we enhance fertilizer selection, thereby increasing yields and enhancing soil health for farmers.