Microsoft Azure AI Fundamentals (AI-900)Describe features of Natural Language Processing (NLP) workloads on AzureEasy
A global technology company receives millions of customer support tickets daily. They want to implement an AI solution that can automatically identify the primary subject or theme of each ticket, such as 'billing inquiry', 'technical support', or 'feature request', without pre-defining specific keywords. Which Azure AI NLP capability is BEST suited for this task?
- ATopic Modeling
- BNamed Entity Recognition (NER)
- CText Summarization
- DSentiment Analysis
Show answer & explanationAnswer & explanation
Correct answer: A. Topic Modeling
Topic Modeling is designed to discover abstract "topics" that occur in a collection of documents. It automatically groups documents that share common themes, making it ideal for categorizing support tickets without explicit keyword rules.
Why the other options are wrong
- B. Named Entity Recognition extracts specific, pre-defined entities like names, locations, or organizations, not overarching themes.
- C. Text Summarization condenses longer text into shorter versions, rather than identifying abstract topics.
- D. Sentiment Analysis determines the emotional tone of text (positive, negative, neutral), not the primary subject.
Topic Modeling
A machine learning technique that automatically identifies abstract 'topics' (themes) that occur in a collection of documents, grouping similar documents together.
- Discovers hidden semantic structures in text.
- Does not require pre-defined keywords for categorization.
- Useful for organizing large document collections.
Memory trick: Topics are like chapters in a big book of text.