Paper Title
A Hybrid Approach to Handle Minority Sampling with Machine Learning Model for Analysing The Diseased Tree

Abstract
Trees are very important in our life. They absorb metric tons of pollution every year and clean the environment. Nowadays, many trees are getting spoiled due to many deadly diseases which are making them sick. In this paper, a disease detection approach has been designed. The disease was identified with the help of machine learning classifiers J48, Naïve Bayes, and SVM to prevent destroying trees. A dataset having diseases was collected from the UCI ML Repository. The dataset was found unbalanced. In order to balance the dataset, a hybrid approach has been developed using the machine learning classifiers J48, Naïve Bayes, and SVM. The results were found efficient at the 8th and 11th iterations. For the 8th iteration, the results of the hybrid technique were 76% accurate for J48, 78.29% for Naive Bayes, and 89.34% for SVM. For the 11th iteration J48 was found 88.92% accurate, Naive Bayes 90.53% accurate, and SVM 97.27% accurate. The proposed approach outperformed all other classifiers. Keywords - Hybrid Approach, J48, Machine Learning ML), Naive Bayes, SVM (Support Vector Machine)