The correct answer is: A. a and b.
The described method would, in general, help restrict the size of the trees. This is because the method only splits tree nodes if the resultant decrease in the sum of squares error exceeds some threshold. This means that the method will only split tree nodes if it is confident that the split will improve the model’s performance. As a result, the method will tend to produce smaller trees than other methods.
The described method also has the potential to affect the performance of the resultant regression/classification model. This is because the method may not always split tree nodes at the optimal location. As a result, the model may not always be as accurate as it could be.
The described method is not computationally infeasible. This is because the method can be implemented using a simple algorithm. The algorithm can be implemented in a variety of programming languages, such as Python, R, and Java.
Here is a more detailed explanation of each option:
- Option a: It would, in general, help restrict the size of the trees. This is because the method only splits tree nodes if the resultant decrease in the sum of squares error exceeds some threshold. This means that the method will only split tree nodes if it is confident that the split will improve the model’s performance. As a result, the method will tend to produce smaller trees than other methods.
- Option b: It has the potential to affect the performance of the resultant regression/classification model. This is because the method may not always split tree nodes at the optimal location. As a result, the model may not always be as accurate as it could be.
- Option c: It is computationally infeasible. This is not true. The described method can be implemented using a simple algorithm. The algorithm can be implemented in a variety of programming languages, such as Python, R, and Java.