Page 118 - DLIS402_INFORMATION_ANALYSIS_AND_REPACKAGING
P. 118

Unit 6: Information Retrieval Model and Search Strategies




            F-measure                                                                                Notes
            Main article: F-score
            The weighted harmonic mean of precision and recall, the traditional F-measure or balanced F-score
            is:

                                     2 . precision . recall
                                 F =
                                     (precision +  recall)
            This is also known as the F  measure, because recall and precision are evenly weighted.
                                  1
            The general formula for non-negative real β is:
                                            .
                                     (1 + β 2 ) (precision . recall)
                                 F  =   β (  2  . precision  + recall)  .
                                  β
            Two other commonly used F measures are the F  measure, which weights recall twice as much as
                                                   2
            precision, and the F  measure, which weights precision twice as much as recall.
                            0.5
            The F-measure was derived by van Rijsbergen (1979) so that F  “measures the effectiveness of retrieval
                                                            β
            with respect to a user who attaches β times as much importance to recall as precision”. It is based on
                                                      1
            van Rijsbergen’s effectiveness  measure E = 1 –   . Their relationship is F = 1 “ E where
                                                   α  +  1 −  α                β
                                                   P    R

                 1
            α =      .
               1 +β 2

            Average Precision
            Precision and recall are single-value metrics based on the whole list of documents returned by the
            system. For systems that return a ranked sequence of documents, it is desirable to also consider the
            order in which the returned documents are presented. Average precision emphasizes ranking relevant
            documents higher. It is the average of precisions computed at the point of each of the relevant
            documents in the ranked sequence:

                                               N
                                              ∑   ( P( r × rel ( ))
                                                     )
                                                           r
                                              r = 1
                                 AveP =
                                        number of relevant documents
            where r is the rank, N the number retrieved, rel() a binary function on the relevance of a given rank,
            and P(r) precision at a given cut-off rank:
                               |{relevant retrieved documents of rank   or less}|
                                                                  r
                          P(r) =
                                                    r
            This metric is also sometimes referred to geometrically as the area under the Precision-Recall curve.



                     The denominator (number of relevant documents) is the number of relevant documents
                     in the entire collection, so that the metric reflects performance over all relevant
                     documents, regardless of a retrieval cut off.






                                             LOVELY PROFESSIONAL UNIVERSITY                                   113
   113   114   115   116   117   118   119   120   121   122   123