Mushroom Poisonous Prediction Based on the Logistic Regression Model

Abstract

Mushroom poisoning is a critical food safety issue that poses a significant threat to public health. In China alone, in 2022, there were 1332 reported cases of mushroom poisoning, leading to 28 deaths. With the existence of numerous toxic mushroom species that can cause fatal consequences, it is crucial to develop a reliable model for predicting mushroom toxicity. In this study, a logistic regression model was developed using a dataset consisting of over 60,000 records with 20 different variables. The model was constructed using the Python programming language and the Scikit-learn package. Logistic regression is a binary classification algorithm that uses a sigmoid function to transform its results into probability values. It relies on relationships between variables to predict a value and subsequently converts it to either a positive or negative outcome, corresponding to two classes. The results of this study indicated that the developed model achieved 100% prediction accuracy with the use of 2000 or more records in the dataset. Therefore, the proposed logistic regression model presents a promising tool for accurately predicting mushroom toxicity and mitigating the risks associated with mushroom poisoning. Further research could focus on expanding the dataset to include additional variables, such as environmental and geographical factors, to improve the model's accuracy and applicability.