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. You need a set of models to…

Question

Exhibit

Options

• A1 = AI Platform, 2 = AI Platform, 3 = AutoML Vision
• B1 = AI Platform, 2 = AI Platform, 3 = AutoML Natural Language
• C1 = AI Platform, 2 = AI Platform, 3 = Cloud Natural Language API
• D1 = Cloud Natural Language API, 2 = AI Platform, 3 = Cloud Vision API

• A
  5% (1)
• B
  5% (1)
• C
  75% (15)
• D
  15% (3)

Explanation

The three enrichment tasks map to the following services: (1) Ticket priority prediction is a custom classification task requiring a custom-trained model → AI Platform. (2) Ticket resolution time prediction is also a custom regression/classification task → AI Platform. (3) Sentiment analysis involves general-purpose natural language understanding with no domain-specific jargon, which is precisely the use case for the pre-built Cloud Natural Language API - no custom model training is needed and it provides sentiment analysis out of the box. AutoML Natural Language (Option B) would require labeling and training a custom model unnecessarily. Cloud Vision API (Options A and D) is for image analysis, not text, making those options invalid for sentiment analysis on support tickets.

No community discussion yet for this question.