Mushroom Quality Prediction using Machine Learning Classification

Abstract

For machine learning applications, classification is the first step in grouping, dividing, categorization and separation of dataset based on the feature vectors. Mushrooms are the most familiar delicious food which is cholesterol free as well as rich in vitamins and minerals. Though nearly 45000 species of mushrooms having known throughout the world, most of them are poisonous and few are lethally poisonous. In this Project we focus on the use of classification Techniques such as Bayes and Functions classifiers to predict the quality of mushroom for its edibility. For performing the experiment we will use a mushroom dataset which is available in UCI Machine Learning Repository, which includes descriptions of hypothetical samples corresponding to 23 species of Gilled Mushrooms in Agaricus and Lepiota family. We will use different Machine Learning algorithms to check the quality of mushroom and analyze which algorithm performs better. For experimentation purpose, we have used WEKA tool and the Mushroom dataset used in our work is downloaded from Mushroom Dataset Datahub. The performance of different techniques is evaluated using different parameters such as accuracy, MAE, and kappa statistics.