precision and recall
(50 minutes to learn)
In pattern recognition and information retrieval, precision (also called positive predictive value) is the fraction of retrieved instances that are relevant, while recall (also called sensitivity) is the fraction of relevant instances that are retrieved (Wikipedia). For instance, if there were 50 relevant documents in a corpus where 20 of the 50 documents were relevant to a user, and an information retrieval (IR) system returned 20 documents, where 6 of the documents were relevant, the recall would be 6/50 = 0.12, and the precision would be 6/20 = 0.3.
-this concept has no prerequisites-
- Understand the definition of precision and recall and why optimizing an information retrieval system solely for precision or recall is a bad idea.
- What is the difference between precision and recall?
- Should you use precision or recall for ranked retrieval sets?
Core resources (read/watch one of the following)
→ Introduction to Information Retrieval
A textbook on information retrieval techniques.
Location: "Evaluation of Unranked Retrieval Sets"
Supplemental resources (the following are optional, but you may find them useful)
Location: Article: Precision and Recall
- F measure is the weighted harmonic mean of the recall and precision and provides a measure of overall quality of an information retrieval system
- create concept: shift + click on graph
- change concept title: shift + click on existing concept
- link together concepts: shift + click drag from one concept to another
- remove concept from graph: click on concept then press delete/backspace
- add associated content to concept: click the small circle that appears on the node when hovering over it
- other actions: use the icons in the upper right corner to optimize the graph placement, preview the graph, or download a json representation