Impact of temperature condition in crop disease analyzing using machine learning algorithm

Abstract

In this study K-Nearest Neighbor (KNN) and Max Voting methods, we compare the accuracy rate and RMSE of the prediction system by using temperature data to predict crop disease. The crop disease is determined by analyzing the Kaggle crop data set, which contains sample sizes of 500 (Group 1 = 250 and Group 2 = 250) based on the temperature conditions. To improve accuracy and reduce RMSE, K-Nearest Neighbor (KNN) and Max Vote methods are used to predict crop diseases based on temperature conditions. Max Voting produced 91% accurate results, while K-Nearest Neighbor (KNN) produced 88% accurate results. According to the Max-voting method, the root mean square error (RMSE) is 3.97, while the K-Nearest Neighbor (KNN) method has an RMSE of 4.34. These two measures are significantly different (P = 0.0356, 0.0421 respectively). For crop disease prediction based on temperature, the Max Voting method performs better than the K-nearest Neighbor (KNN) method.

Keywords

• Crop disease prediction
• Novel max voting classifier
• K-nearest neighbor
• Accuracy rate
• RMSE
• Temperature condition