This project applied Decision Tree analysis to a new dataset, aiming to provide hands-on experience in data analysis. I selected a wine quality dataset, where the outcome variable was the wine quality rating, and analyzed how features like alcohol content and acidity influenced the predictions. By testing various minimum split values, I determined that a split of 30 optimized model accuracy. I also identified the nodes with the highest and lowest probabilities, gaining insights into which factors most impacted wine quality predictions. Lastly, I used the trained model to predict wine quality for new data points, demonstrating the practical application of the model. Through this project, I enhanced my understanding of Decision Trees, model tuning, and data-driven decision-making.
Connect and innovate with an elite information systems program
