passdrill
AI & LLM Engineering · RAG & Embeddings · Card 009/012 medium

A retrieval system is evaluated on a query for which there are 10 truly relevant passages somewhere in the corpus. The system returns 20 passages for that query, and 8 of those 20 are among the 10 truly relevant passages. What are this query's recall@20 and precision@20?

  1. Recall@20 = 8/10 = 0.8, because 8 of the 10 relevant passages were retrieved; precision@20 = 8/20 = 0.4, because only 8 of the 20 returned passages were relevant
  2. Recall@20 = 8/20 = 0.4, because 8 of the 20 returned passages were relevant; precision@20 = 8/10 = 0.8, because 8 of the 10 relevant passages were retrieved
  3. Recall@20 and precision@20 are both 8/18 = 0.44, since the two missed relevant passages and the twelve irrelevant returned passages should be pooled into a single combined denominator
  4. Recall@20 cannot be computed from the numbers given because it requires knowing the total size of the corpus, whereas precision@20 can be computed and equals 8/20 = 0.4
Next card → Shuffle