A healthcare analytics company wants to segment patients into groups that have similar risk factors to develop personalized treatment plans. The company has a dataset that includes patient health reco

Question

Options

• AUse the Amazon SageMaker AI XGBoost algorithm. Set max_depth to control tree complexity for
• BUse the Amazon SageMaker k-means clustering algorithm. Set k to specify the number of
• CUse the Amazon SageMaker AI DeepAR algorithm. Set epochs to determine the number of
• DUse the Amazon SageMaker AI Random Cut Forest (RCF) algorithm. Set a contamination

• A
  10% (3)
• B
  81% (25)
• C
  6% (2)
• D
  3% (1)

Explanation

Patient segmentation into similar-risk groups is an unsupervised clustering task - there are no predefined labels. The Amazon SageMaker k-means algorithm is purpose-built for clustering and uses the hyperparameter k to specify how many clusters (groups) to create, which directly addresses the requirement to 'determine the number of groups by using hyperparameters.' XGBoost (A) is a supervised learning algorithm for classification or regression, not clustering. DeepAR (C) is a time-series forecasting algorithm. Random Cut Forest (D) is an anomaly detection algorithm used to find outliers, not to segment patients into balanced groups.

No community discussion yet for this question.