K-Means with Epsilon Neighborhood Technique on Crop Yield Prediction

Abstract

Agriculture has a significant impact on the economy of developing countries.Global changes in climatic conditions and agricultural investment costs are the mainobstacles faced by small farmers. The proposed work aims to design a predictive model that provides farmingplans to high-yielding farmers using data mining techniques. Data mining techniques extract hidden knowledge through data analysis, unlike statistical approaches. The dataset is collected from the agricultural department. Variousdata mining techniques, such as Decision Trees, Naïve Bayes, Density-Based clustering, and K Means clustering, were applied to the agricultural data for better crop prediction.In this paper, K-Means with Epsilon Neighborhood-based prediction model is developed. The result of the proposed work is an accurate prediction of crop yield. The final rules extracted from this work are helpful for farmers to make proactive and informed decisions before harvest.