Published Paper Details:

In agriculture, the integration of machine learning has been a long-standing aspiration, resulting in significant advancements. While machine learning models have been developed for crop and yield predictions, traditional algorithms like decision trees often fall short of delivering the desired accuracy. This paper introduces an accessible and user-friendly solution for crop recommendations and yield predictions. Users provide inputs such as temperature, humidity, soil pH, and rainfall. To enhance accuracy, a hybrid approach using K-nearest neighbor (KNN) and Random Forest (RF) algorithms is employed. The K-nearest neighbor (KNN) algorithm achieves an impressive accuracy rate of 98%. Additionally, the Random Forest (RF) algorithm attains a commendable 96% accuracy by aggregating multiple decision trees. These high accuracy rates signify the system's potential to empower farmers with data-driven insights for crop selection and yield projections. Furthermore, the user-friendly interface promises broader adoption within the agriculture sector, catering to users with varying levels of technical proficiency. To strengthen the system's credibility, transparency regarding data sources and quality is imperative. Utilizing accurate and relevant data for reliable predictions. In summary, this paper presents a promising solution for informed decision-making in agriculture, combining crop recommendations and yield predictions. Acknowledging the limitations of traditional approaches, it capitalizes on the strengths of K-nearest neighbor and Random Forest algorithms.