AI-102 Key Phrase Extraction Practice Question
You are a developer at a large financial institution. The compliance team needs to automatically analyze quarterly earnings call transcripts to extract forward-looking statements (e.g., 'we expect revenue to grow') and flag any that are overly optimistic or lack necessary disclaimers. The transcripts are stored as text files in Azure Blob Storage. You need to design a solution using Azure AI Language services that meets the following requirements: 1) Extract all forward-looking statements from each transcript. 2) For each statement, determine if it contains optimistic language (e.g., 'strong growth', 'excellent performance') and if it includes a disclaimer (e.g., 'this is a forward-looking statement'). 3) Output a structured JSON file per transcript with the statements, optimism score, and disclaimer presence. 4) Minimize development effort and avoid custom machine learning model training. Which approach should you take?
⚠ Common exam trap
The trap is that candidates assume custom models (NER and text classification) are necessary for detailed extraction and classification tasks, but the requirement explicitly prohibits custom ML training. Candidates may overlook that prebuilt features like key phrase extraction and sentiment analysis, combined with simple logic, can approximate the required functionality with less effort.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Use key phrase extraction to identify important phrases, then run sentiment analysis on each sentence to detect optimism.
It uses prebuilt Azure AI Language features (key phrase extraction and sentiment analysis) that require no custom ML model training, meeting the requirement to avoid custom training. Key phrase extraction can identify potential forward-looking phrases, and sentiment analysis on each sentence can provide a per-statement optimism score. Disclaimer presence can be inferred by checking for specific phrases like 'forward-looking statement' using simple text matching or by incorporating the sentiment analysis result for that sentence. This approach minimizes development effort while providing structured output per transcript.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Use the prebuilt named entity recognition (NER) to identify entities related to financial terms, then apply sentiment analysis to the entire transcript to determine overall optimism.
Why it's wrong here
Using prebuilt NER and sentiment analysis on the entire transcript does not extract individual forward-looking statements or provide per-statement optimism and disclaimer detection. It only gives overall sentiment, which does not meet the detailed requirement.
- ✗
Build a custom NER model to extract forward-looking statements, then use a custom text classification model to classify each extracted statement for optimism and disclaimer presence.
Why it's wrong here
This option requires building custom NER and custom text classification models, which directly contradicts the requirement to avoid custom machine learning model training. It also involves significant development effort.
- ✗
Use custom question answering to create a knowledge base of typical forward-looking statements and query the transcript for matches.
Why it's wrong here
Custom question answering requires creating a knowledge base of typical forward-looking statements, which is a custom effort and not suitable for extracting statements from arbitrary transcripts. It does not provide per-statement optimism or disclaimer analysis.
- ✓
Use key phrase extraction to identify important phrases, then run sentiment analysis on each sentence to detect optimism.
Why this is correct
This approach uses prebuilt key phrase extraction to identify important phrases, then runs sentiment analysis on each sentence to detect optimism. Disclaimer presence can be checked with simple pattern matching. It avoids custom ML training and minimizes development effort, meeting all requirements.
Quick reference
Azure Blob Storage Tier Comparison
| Tier | Storage Cost | Retrieval Cost | Latency | Use Case |
|---|---|---|---|---|
| Hot | Highest | Lowest | Immediate | Active data, frequent reads |
| Cool | Lower | Higher | Immediate | Data accessed < once / month |
| Cold | Lower still | Higher | Immediate | Data accessed < once / quarter |
| Archive | Lowest | Highest + rehydration delay | Hours | Long-term compliance retention |
Go deeper
Related to this question
About these practice questions
This AI-102 question is part of Courseiva's 945-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-102 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-102 exam.