This Streamlit-based project aims to predict lung cancer using a Decision Tree Classifier. It begins with dataset upload, preprocessing (mapping categorical values), and splitting into training, test, and output sets. The model is trained and evaluated on accuracy and mean absolute error metrics. Predictions are made on both test and output datasets, with results displayed alongside actual values for comparison. Users can upload a CSV, view dataset details, and save predicted outputs to a CSV file. The interface offers insights into feature importance for potential clinical applications.
This Streamlit-based project aims to predict lung cancer using a Decision Tree Classifier. It begins with dataset upload, preprocessing (mapping categorical values), and splitting into training, test, and output sets. The model is trained and evaluated on accuracy and mean absolute error metrics. Predictions are made on both test and output datasets, with results displayed alongside actual values for comparison. Users can upload a CSV, view dataset details, and save predicted outputs to a CSV file. The interface offers insights into feature importance for potential clinical applications.