Design of K-Means Clustering & SVM based Models for Soil Fertility Classification

A fertile soil is essential for healthy plant growth and quality food. This work aims to classify samples into more fertile soil and less fertile soil. It has been carried out using the soil fertility data of Kumaun, Uttarakhand. The climate and soil quality is very promising in this region of India. Here the soil has plenty of nutrients and the best pH value resulting in an outstanding fertility. Moreover, in this area, very favorable temperature and moisture conditions make the agriculture economically sound. In this work, machine learning techniques, k-means clustering and Support Vector Machine, are used in MATLAB to create models for classifying the soil fertility using the potassium concentration and pH of soil as inputs. The trained models are also tested against the test data to compare the classification accuracy of the developed models.