Decision Tree is a supervised learning algorithm that uses a tree-like model of decisions to predict an outcome, Random Forest is an ensemble of Decision Trees, using bagging technique to decrease overfitting, Neural Network is a supervised learning algorithm that is inspired by the structure and function of the human brain, used for image recognition, speech recognition, natural language processing and many other tasks.
In this project, we used three machine learning algorithms above to classify iris species and income classes and compared the performance of the three algorithms in two datasets.
-
The Iris dataset has simple features and small data volume, and both decision trees and neural networks perform well on this dataset with high accuracy rates of over 90%. the Income dataset has many features and large data volume, and the overall dataset is more complex, and both decision trees and neural networks achieve more excellent results.
-
In the Iris dataset experiments, the model was trained and tested in a very short time due to the small amount of data. In the Income dataset experiments, the difference in time between the two is more obvious with larger data volume and more features.
-
Decision trees, as strongly interpretable models, perform very well on the Iris dataset. Neural networks are much less interpretable than decision trees due to their algorithmic characteristics.