- Easy to understand and interpret, perfect for visual representation. This is an example of a white box model, which closely mimics the human decision-making process.
- Can work with numerical and categorical features.
- Requires little data preprocessing: no need for one-hot encoding, dummy variables, and so on.
- Non-parametric model: no assumptions about the shape of data.
- Fast for inference.
- Feature selection happens automatically: unimportant features will not influence the result. The presence of features that depend on each other (multicollinearity) also doesn't affect the quality.
- It tends to overfit. This usually can be mitigated in one of three ways:
- Limiting tree depth ...