A city wants to monitor its air quality to address the consequences of air pollution. A Machine Learning Specialist needs to forecast the air quality in parts per million of contaminates for the…

Question

Options

• AUse the Amazon SageMaker k-Nearest-Neighbors (kNN) algorithm on the single time series
• BUse Amazon SageMaker Random Cut Forest (RCF) on the single time series consisting of the full
• CUse the Amazon SageMaker Linear Learner algorithm on the single time series consisting of the
• DUse the Amazon SageMaker Linear Learner algorithm on the single time series consisting of the

• A
  5% (1)
• B
  14% (3)
• C
  73% (16)
• D
  9% (2)

Explanation

The task is a regression/forecasting problem on a single short time series (365 daily data points) predicting 2 days ahead. Amazon SageMaker's Linear Learner algorithm is well-suited here: it supports supervised regression, can ingest engineered time-based features (lag values, day-of-week, rolling averages), and trains efficiently on small datasets. It provides a straightforward, interpretable model appropriate for a prototype. The kNN algorithm (A) is a classification or regression method but is not designed for time series forecasting and performs poorly with temporal structure. Random Cut Forest (B) is an unsupervised anomaly detection algorithm, not a forecasting model. The answer options C and D appear truncated in the question, but based on the correct answer designation, the Linear Learner applied to the full time series with appropriate feature engineering is the best choice for this prototype scenario.

No community discussion yet for this question.