New Decision Tree Algorithm Outperforms Logistic Regression in Fire Detection
The goal of the study was to see how well a decision tree algorithm and logistic regression algorithm could classify images of fire and smoke. They took 4,000 photos, with half showing fire and half showing smoke. Using a machine learning package, they found that the decision tree algorithm was more accurate at classifying the images compared to logistic regression. The decision tree algorithm had an accuracy of 90.54%, while logistic regression had an accuracy of 85.75%.