Abstract: Accurately identifying edible and toxic mushrooms is essential to preventing foodborne illnesses, as misclassification can lead to severe health risks. Traditional classification methods often depend on expert judgment, which can be subjective, time-consuming, and prone to errors. This study investigates the use of machine learning to automate the classification of mushrooms into edible and poisonous categories. Leveraging the UCI Mushroom Dataset, which includes features such as cap shape, color, gill spacing, and habitat, we evaluate three machine learning models: Decision Trees, Random Forests, and Logistic Regression. The findings demonstrate that these models achieve high accuracy, proving their effectiveness in mushroom classification. To enhance model performance, preprocessing techniques such as feature selection and handling class imbalances are applied. The results highlight the potential of machine learning in improving food safety, assisting foragers, and supporting agricultural applications. Future work could explore deep learning for image-based classification and incorporate environmental factors to refine real-time decision-making systems.
Keywords: Machine Learning, Mushroom Classification, Decision Tree, Random Forest, Logistic Regression.
Copyright Form