Edible and Poisonous Mushroom Classification Using Machine Learning

Abstract

Mushrooms are widely consumed as a nutritious food source due to their high protein, vitamin, and mineral content. However, several mushroom species are highly poisonous and resemble edible varieties, making manual identification difficult even for experienced collectors. Consumption of poisonous mushrooms can lead to severe health complications and even death. This research presents a machine learning-based mushroom classification system that accurately distinguishes edible mushrooms from poisonous ones using their physical characteristics. The proposed system employs supervised machine learning algorithms trained on a labeled mushroom dataset containing features such as cap shape, cap color, odor, gill size, stalk characteristics, and habitat. Data preprocessing techniques including missing value handling, categorical feature encoding, and feature scaling are applied before model training. Various classification algorithms such as Decision Tree, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Naïve Bayes are evaluated based on accuracy, precision, recall, and F1-score. Experimental results indicate that ensemble learning methods, particularly Random Forest, achieve superior classification performance with accuracy exceeding 99%. The developed system offers a reliable, efficient, and cost-effective solution for mushroom identification and can support farmers, researchers, food industries, and consumers in preventing accidental poisoning.

Keywords