Approximate Search: Recall@k vs Exact Search and the Speed Trade-off
This lesson is part of the course
In this lesson
You compare an approximate nearest-neighbour (ANN) index with exact search for six test queries, compute recall@5 per query and on average, then pick the fastest index setting that reaches a 95% recall target.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Embeddings as Vectors: Length, Normalisation & the L1 Norm Locked
- Dot Product vs Cosine Similarity with SUMPRODUCT Locked
- Top-k Retrieval: RANK.EQ, LARGE and INDEX/MATCH Locked
- Euclidean Distance vs Cosine: When Do the Rankings Disagree? Locked
- Chunking: Tokens per Chunk, Number of Chunks & Embedding Cost Locked
- Approximate Search: Recall@k vs Exact Search and the Speed Trade-off Locked
- Hybrid Search: Blending Keyword and Vector Scores Locked
- RAG Retrieval Evaluation: Hit Rate@k, MRR & a Chart Locked