Answer-first summary for fast verification
Answer: Include a time-based feature extraction stage and use an ARIMA model.
Time series forecasting requires considering temporal dependencies, which can be addressed by including a time-based feature extraction stage and using a model like ARIMA that accounts for temporal patterns. Ignoring temporal order or using models that do not account for time dependencies can lead to inaccurate forecasts.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
Consider a scenario where you need to develop a Spark ML pipeline for a time series forecasting task. The pipeline includes feature extraction, model training, and evaluation. Describe how you would structure this pipeline, including the specific stages and components you would use. Additionally, discuss any challenges specific to time series forecasting that you might encounter and how you would address them.
A
Use a simple linear regression model without considering temporal dependencies.
B
Include a time-based feature extraction stage and use an ARIMA model.
C
Ignore the temporal order of data and use a random forest model.
D
Use only the most recent data point for forecasting.
No comments yet.