Crop Selection Model by Implementing the SVM and K-mean Clustering on Soil

Abstract

This is a major challenge for the farmers to decide the crop according to the fertility of soil. Production of crop is purely based on the minerals present in the soil. Each crop required deferent types of soil for its growth. If crop selection will not be done according to the characteristics of crop it will give long term loss to the farmers. The proposed method used two different algorithms, which are SVM (Support Vector machine) and the K-means clustering for the selection of crops based on soil test report. The SVM (Support Vector machine) and the K-means clustering, both the algorithms are used for classifying the datasets in two the different cluster so that we can fetch batter result from those cluster. In SVM the implementation little bit difficult then the k-mean clustering but it found the batter relation and the accuracy then the k-means clustering. SVM gives 65.04% accuracy while K-means clustering gives the accuracy of 56.08%. So we can implement the crop selection model by implementing the SVM on the soil test datasets. It will give the batter results if the data set consist of a huge amount of the data. It also contributes to select the crops, which are suitable for market conditions. The proposed method also used Weather Forecasting module.In Weather Forecasting module, we check the weather condition for any location in the short term or long term. It helps the farmers to select the crops and fertilizers, which are suitable for weather condition