F1

Glossaries | EDRM Glossary | Grossman-Cormack Glossary | Grossman-Cormack Glossary index | Submit a Definition |

Definition

  • The of and , often used in studies as a measure of the effectiveness of a search or review effort, which accounts for the tradeoff between Recall and Precision. In order to achieve a high F1 score, a search or review effort must achieve both high Recall and high Precision. 1
  • One form of van Rijsbergen’s F formula for combining precision and recall into a single number to make it easier to compare the information retrieval accuracy of different systems. F1 is the weighted harmonic mean of precision and recall = 2 * precision * recall / (precision + recall). 2

See also


Notes

  1. Maura R. Grossman and Gordon V. Cormack, EDRM page & The Grossman-Cormack Glossary of Technology-Assisted Review, with Foreword by John M. Facciola, U.S. Magistrate Judge2013 Fed. Cts. L. Rev. 7 (January 2013).
  2. Herb Roitblat, Predictive Coding Glossary.
Please complete the following to download file

Your Name *

Your Email *

Your Organization

Organization Type
 Association Corporation Government Law Firm Provider

Joining EDRM
 Please contact me about EDRM membership

Subscribe me to the mailing list

The reciprocal of the average of the reciprocals of two or more quantities. If the quantities are named a and b, their Harmonic Mean is 2 / ( ( 1 / a ) + ( 1 / b ) ). In information Retrieval, F1 is the Harmonic Mean of Recall and Precision. The Harmonic Mean, unlike the more common arithmetic mean (i.e., average), falls closer to the lower of the two quantities. As a summary measure, a Harmonic Mean may be preferable to an arithmetic mean because a high Harmonic Mean depends on both high Recall and high Precision, whereas a high arithmetic mean can be achieved with high Recall at the expense of low Precision, or high Precision at the expense of low Recall.

Maura R. Grossman and Gordon V. Cormack, EDRM page & The Grossman-Cormack Glossary of Technology-Assisted Review, with Foreword by John M. Facciola, U.S. Magistrate Judge2013 Fed. Cts. L. Rev. 7 (January 2013).

The fraction of Relevant Documents that are identified as Relevant by a search or review effort.

Maura R. Grossman and Gordon V. Cormack, EDRM page & The Grossman-Cormack Glossary of Technology-Assisted Review, with Foreword by John M. Facciola, U.S. Magistrate Judge2013 Fed. Cts. L. Rev. 7 (January 2013).

The fraction of Documents identified as Relevant by a search or review effort, that are in fact Relevant. Also referred to as Positive Predictive Value.

Maura R. Grossman and Gordon V. Cormack, EDRM page & The Grossman-Cormack Glossary of Technology-Assisted Review, with Foreword by John M. Facciola, U.S. Magistrate Judge2013 Fed. Cts. L. Rev. 7 (January 2013).

The science of how to find information to meet an Information Need. While modern Information Retrieval relies heavily on computers, the discipline predates the invention of computers.

Maura R. Grossman and Gordon V. Cormack, EDRM page & The Grossman-Cormack Glossary of Technology-Assisted Review, with Foreword by John M. Facciola, U.S. Magistrate Judge2013 Fed. Cts. L. Rev. 7 (January 2013).

Leave a Reply