Answer-first summary for fast verification
Answer: Develop a simple heuristic (e.g., based on z-score) to label the machines’ historical performance data. Test this heuristic in a production environment.
The correct answer is C. The best initial approach is to develop a simple heuristic, such as one based on z-scores, to label the machines' historical performance data and test this heuristic in a production environment. This approach is in line with Google's machine learning guidelines, which suggest starting with simple solutions to obtain a baseline and then deciding if more sophisticated models are needed. Implementing and testing a heuristic first provides quick and inexpensive insights while minimizing the risk associated with deploying untested models directly into production.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You work on an operations team at an international company that manages a large fleet of on-premises servers located in a few data centers around the world. Your team collects monitoring data from the servers, including metrics such as CPU and memory consumption, to ensure smooth operations. When an incident occurs on a server, it is your team's responsibility to fix it promptly to maintain system uptime. So far, incident data has not been properly labeled, making it challenging to perform predictive maintenance. Your management team has charged you with building a predictive maintenance solution that can use the monitoring data from the virtual machines (VMs) to detect potential failures and generate alerts for the service desk team. Given these requirements, what should be your first step?
A
Train a time-series model to predict the machines’ performance values. Configure an alert if a machine’s actual performance values significantly differ from the predicted performance values.
B
Implement a simple heuristic (e.g., based on z-score) to label the machines’ historical performance data. Train a model to predict anomalies based on this labeled dataset.
C
Develop a simple heuristic (e.g., based on z-score) to label the machines’ historical performance data. Test this heuristic in a production environment.
D
Hire a team of qualified analysts to review and label the machines’ historical performance data. Train a model based on this manually labeled dataset.