AI-102 Implement generative AI solutions Practice Question
You operate a RAG assistant on Azure OpenAI Service that answers questions over a product catalog. The index contains thousands of items, and users complain that answers mix details from unrelated products. You need retrieval to consider semantic meaning and handle paraphrased queries while returning only the most relevant items. What should you implement?
⚠ Common exam trap
The trap here is assuming that adding filters or scoring boosts delivers semantic understanding, when only embeddings capture meaning.
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
✓
Vector search using embeddings generated by an Azure OpenAI embedding model
Embedding-based vector search maps both the query and the catalog items into a semantic space, so similarity reflects meaning rather than exact wording. Paraphrased questions then retrieve the right products, and ranking by vector distance keeps unrelated items out of the context. Keyword, facet, and scoring-profile approaches remain term- or attribute-driven and cannot perform semantic matching.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Faceted navigation that filters results by product category
Why it's wrong here
Facets narrow results to explicit attribute values and are excellent for guided browsing, but they require the query to specify or select a facet. They do not measure semantic similarity, so a paraphrased question would still be matched lexically and could return items from the wrong category. Facets complement retrieval rather than replace semantic ranking.
- ✗
Scoring profiles that boost documents containing a specific field value
Why it's wrong here
Scoring profiles adjust ranking weights when certain fields or values are present, which helps promote known items. They still operate on the keyword match pipeline and cannot infer meaning from a paraphrased question. Because relevance remains term-based, unrelated products with matching words could still be retrieved, leaving the core complaint unaddressed.
- ✓
Vector search using embeddings generated by an Azure OpenAI embedding model
Why this is correct
Vector search compares the embedding of the query against embeddings of the catalog items, so semantically similar content ranks highly even when wording differs. This handles paraphrased queries and sharpens relevance, which reduces the chance of pulling details from unrelated products. It is the appropriate retrieval mode for the described semantic matching requirement.
- ✗
Keyword search using the Azure AI Search simple query syntax
Why it's wrong here
Keyword search matches literal terms, so a paraphrased query using different vocabulary than the catalog text may retrieve nothing relevant. It also cannot rank by conceptual similarity, which is what distinguishes related products from unrelated ones. This approach would leave the mixing problem largely unsolved because recall depends on exact term overlap between query and documents.
Go deeper
Related to this question
About these practice questions
This AI-102 question is part of Courseiva's 761-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 and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Microsoft exam blueprint
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.