Advantage
- It is very easy, effective and simple.
- It can handle both categorical and numeric data very efficiently as compared to other algorithms.
- Missing values present in the data set does not affect the decision tree.
- Preprocessing of data is not required.
- Results that the decision tree generate does not require any prior knowledge of statistical or mathematics.
Disadvantages
- If data is not discretized properly, then it will give an inaccurate result.
- It is sometimes unstable and cannot be reliable because of alteration in data which causes a bad structure of the decision tree which affects the model.
- It works very badly in the case of regression as it does not support data with too much variation.