Learning from Data

Question 1

What are the two types of Learning?

Answer 1

Supervised Learning and Unsupervised Learning.

Question 2

What is supervised learning?

Answer 2

The most common form of machine learning, it requires that training data be labelled and attempts to learn a mapping function to go from input features to output labels.

Question 3

What is unsupervised learning?

Answer 3

A Machine Learning method that attempts to “make sense” of the data and does not have the associated labels for data points.

Question 4

What are the parameters in a supervised learning model?

Answer 4

These are the values that change as the model learns from the data. In a linear regression model, this would be the beta values.

Question 5

What are hyperparameters?

Answer 5

A parameter not learned directly from the data but relates to implementation.

Question 6

How is predicition quality measured in a supervised learning model?

Answer 6

A loss function is used; this provides some quantitative measure of how close the predictions are to the actual values.

Question 7

What are the types of supervised learning?

Answer 7

Regression and Classification.

Question 8

What is a regression model used for?

Answer 8

Regression is used for modeling any continuos target or outcome such as loss, revenue, time etc

Question 9

What is a classification model used for?

Answer 9

Classification is used to identify which set of categories an observation belongs to.