Forecasting Heart Attack with Decision Tree

Our Unit Lead, Daniel Baldini, presented a talk titled “Forecasting Heart Attack with Decision Tree” at the most recent Friday Talk. He explained the fundamental concepts that help people understand the results of the Decision Tree model and classification models in general. He used a public database to forecast heart attacks using the decision tree algorithm.

During the talk, Daniel discussed comprehending the problem on the business level, understanding the data available, separating classification from regression, delved into the issue of how to create the confusion matrix and continued by explaining why the cutoff is essential. He also discussed the importance of understanding certain metrics for evaluating the model, testing and training the Split technique, playing with the Decision Tree algorithm, and exploring some philosophical questions. In the end, he checked the solution’s results with the public.

If you are interested in this topic and would like to find out more about data science problems and other possible algorithms to solve them, watch the full recording of this insightful Friday Talk.