Edible Mushroom Identification Using Machine Learning
K. Kousalya, B. Krishnakumar, S. Boomika, N. Dharati, N. Hemavathy
Abstract
Mushrooms are fungi, and there are many different types of fungi, including moulds and crusts, but there has been minimal research on whether they are edible or dangerous. Learning machine will be able to classify if a mushroom is poisonous or not using data mining as one of the approaches for obtaining computer-assisted knowledge. Naive Bayes, Decision Tree (C4.5), Support Vector Machine (SVM), and Logistic Regression are four comparison of the finest categorization techniques in data mining currently available. The research approach employed was an experiment with a WEKA-assisted tool that was used to compare the four algorithms. The Lepiota mushroom and Agaricus data were used to conduct the testing. Source of the dataset used is collected from the Kaggle website. The C4.5 algorithm, however, has the maximum level of accuracy compared to the other algorithms, according to the results of the testing, from the speed aspect it is noted that the C4.5 algorithm is faster than the other algorithms. This algorithm results in showing maximum accuracy of 93.34%