AWS Certified Machine Learning Study Guide

Assessment Test

1. You are building a supervised ML model for predicting housing prices in the United States. However, you notice that your dataset has a lot of highly correlated features. What are some methods you can use to reduce the number of features in your dataset? (Choose all that apply.)
   1. Use principal component analysis to perform dimensionality reduction.
   2. Add an L2 regularization term to your loss function.
   3. Add an L1 regularization term to your loss function.
   4. Add an L3 regularization term to your loss function.
2. Which of the following is an unsupervised learning algorithm useful with tabular data?
   1. K-nearest neighbors
   2. K-means clustering
   3. Latent Dirichlet Allocation (LDA)
   4. Random forest
3. Which of the following ML instance types is ...