Criteria for splitting :

[Venkat_Subramani]

Putting it together - Decision trees | Coursera

Forward to 4:55

Why was Whiskers Feature chosen on RHS , why can’t we still use Face Shape as it picked for left side splitting which had higher Information Gain.

[Wendy]

Because on the RHS, Whiskers had the highest information gain. You calculate information gain for each node individually.

[Venkat_Subramani]

Thanks Wendy.
I have another qn, Wonder if I should create a new thread or I can continue on same thread to ask qns on this weeks videos?

---

Regression Trees (optional) - Decision trees | Coursera

I am lost on the benefits of predicting a number using decision tree.
Could we not use Linear or perhaps softmax Logistic regression ? I am seeing trees, but I am missing the forest here.

[Wendy]

@Venkat_Subramani

I am seeing trees, but I am missing the forest here.

I’m afraid I don’t have any great insights about when it would make sense to choose a decision tree vs linear regression. Decision trees are popular - at least in places like Kaggle. I suspect this is partly because they make it easier to visualize what’s going on inside the model.

There’s further discussion about this topic on this thread:

Deciding when to use a Decision Tree Model

Hi! I’ve been reading some further material around Tree based models after the first 2 videos: When and Why Tree-Based Models (Often) Outperform Neural Networks | by Andre Ye | Towards Data Science Curious to understand a) Is there a general rule of thumb as to when one should use a decision tree vs. another model e.g. Neural Network? b) Would it be advised to implement both models and compare training sets? c) If yes - then is there any value in amalgamating the two models e.g. A Decision T…

As far as your first question, in general, it’s better to make a new post for new questions. That makes it easier for mentors to notice your post, which means you’ll get a quicker response. It also can make it easier for learners to find the post in the future if they have the same question.