Google Professional Machine Learning Engineer
Get started today
Ultimate access to all questions.
Ultimate access to all questions.
You are designing an architecture with a serverless ML system to enrich customer support tickets with informative metadata before they are routed to a support agent. The system requires models that can predict ticket priority, predict ticket resolution time, and perform sentiment analysis to assist support agents in making strategic decisions when processing support requests. Since tickets will not include any domain-specific terms or jargon, you do not need custom models. The proposed architecture flow involves a user submitting a ticket to Firebase, which triggers a Cloud Function. The Cloud Function will then call three different endpoints for enrichment before updating the Firebase real-time database and creating a ticket in the helpdesk platform using a RESTful API. Which endpoints should the Enrichment Cloud Functions call?
Explanation:
Option C is the correct answer. The architecture requires models to predict ticket priority, predict ticket resolution time, and perform sentiment analysis. Given the needs and the fact that tickets will not contain domain-specific terms or jargon, you do not need custom models. Therefore, for ticket priority (categorical) and resolution time (regression), AI Platform endpoints are suitable. For sentiment analysis, the Cloud Natural Language API is sufficient, as it can handle sentiment analysis without the need for custom training. Thus, the correct set of endpoints is: 1 = AI Platform, 2 = AI Platform, 3 = Cloud Natural Language API.