You are working for a company that does consultancy work on AI. Your first task

Question

You are working for a company that does consultancy work on AI. Your first task is to provide the expert advice on which machine learning technique is best suited for the scenarios below. Your answer should be backed up by an argument as to why the machine learning technique selected is best suited for that task.

Task 1: Involves around 100,000 training examples in a 4-dimensional feature space. The aim is to classify 100 test examples.

Task 2: You are going to develop a classifier to recommend which children should be assigned to special education classes in kindergarten. The classifier has to be justified to the board of education before it is implemented.

Explanation / Answer

Answer for Task 1:
Nearest Neighbors is a good choice. The dimensionality is low and so
appropriate for KNN. For KNN, training is very fast and since there are
few classifications, the fact that this will be slow does not matter. With 1
million training examples, neural net and SVM will be extremely expensive
to train. Naive Bayes is plausible on computational grounds but likely to
be less accurate than KNN

Answer for Task 2:
A Decision Tree is a good choice since the resulting classifier will need to
be understandable to humans.

Related Questions

Navigate

• Browse (All)
• Browse Y
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.