{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}


84 evaluaon idea if locally precision increases with

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: CJ van Rijsbergen, Informa)on Retrieval Introduc)on to Informa)on Retrieval 16 Sec. 8.4 Evalua)ng ranked results   Evalua)on of ranked results:   The system can return any number of results   By taking various numbers of the top returned documents (levels of recall), the evaluator can produce a precision ­ recall curve 17 18 3 Introduc)on to Informa)on Retrieval Sec. 8.4 A precision ­recall curve Introduc)on to Informa)on Retrieval Sec. 8.4 Averaging over queries   A precision ­recall graph for one query isn’t a very sensible thing to look at   You need to average performance over a whole bunch of queries.   But there’s a technical issue:   Precision ­recall calcula)ons place some points on the graph   How do you determine a value (interpolate) between the points? 19 Introduc)on to Informa)on Retrieval Sec. 8.4 Interpolated precision 20 Introduc)on to Informa)on Retrieval Sec. 8.4 Evalua)on   Idea: If locally precision increases with increasing recall, then you should get to count that…   So you take the max of precisions to right of value ...
View Full Document

{[ snackBarMessage ]}