Design and Implement Intelligent Search DP-800 Practice Questions (Page 8)

Questions 36–40

1. 36foundation · easy

   What does the VECTOR_NORMALIZE function do to a vector?

   • AIt reverses the order of elements
   • BIt rounds each element to the nearest integer
   • CIt scales the vector to have a length of 1
   • DIt converts the vector to a different data type
   Reveal Solution Discussion Select an answer first
2. 37foundation · easy

   Which SQL Server feature enables fast keyword-based searches over character data in tables, supporting features like stemming, thesaurus, and stopwords?

   • AIn-Memory OLTP
   • BColumnstore Index
   • CFull-Text Search
   • DJSON Index
   Reveal Solution Discussion Select an answer first
4. 38expert · hard

   A job board runs hybrid search on Azure SQL Database combining full-text and vector search. The full-text search returns highly relevant results for exact skill matches, while vector search finds semantically related job descriptions. The team notices that the fused results are dominated by the vector search scores, and exact skill matches are being pushed down. What is the most likely cause and fix?

   • AThe full-text query is missing the FREETEXT predicate; add it to improve full-text ranking.
   • BThe vector index is using cosine distance instead of Euclidean; change to Euclidean to reduce vector dominance.
   • CThe full-text index is not being used; switch to CONTAINSTABLE to get proper relevance scores.
   • DThe RRF constant k is too low, causing high vector scores to dominate; increase k to balance the ranking.
   Reveal Solution Discussion Select an answer first
5. 39expert · hard

   A team has implemented hybrid search with RRF. They notice that the search results are heavily biased toward the full-text component, and relevant vector-only matches are ranked too low. They want to rebalance the influence of the two components. What should they adjust?

   • AUse a different distance metric for the vector search.
   • BIncrease the number of results returned by the vector search.
   • CRemove the full-text component entirely.
   • DIncrease the RRF constant (k) for the vector component.
   Reveal Solution Discussion Select an answer first
6. 40foundation · easy

   What is the purpose of Reciprocal Rank Fusion (RRF) in hybrid search?

   • ATo combine multiple ranked result lists into a single ranking
   • BTo create a full-text index
   • CTo normalize vectors to unit length
   • DTo calculate the distance between vectors
   Reveal Solution Discussion Select an answer first