In statistics, the F1 score (also F-score or F-measure) is a measure of a test's accuracy. It considers both the precision p and the recall r of the test to compute the score: p is the number of correct results divided by the number of all returned results and r is the number of correct results divided by the number of results that should have been returned. The F1 score can be interpreted as a weighted average of the precision and recall, where an F1 score reaches its best value at 1 and worst score at 0.
The traditional F-measure or balanced F-score (F1 score) is the harmonic mean of precision and recall:
The general formula for positive real β is:
The formula in terms of Type I and type II errors:
Two other commonly used F measures are the measure, which weights recall higher than precision, and the measure, which puts more emphasis on precision than recall.
The F-measure was derived so that "measures the effectiveness of retrieval with respect to a user who attaches β times as much importance to recall as precision". It is based on Van Rijsbergen's effectiveness measure
Their relationship is where .
The F-score is often used in the field of information retrieval for measuring search, document classification, and query classification performance. Earlier works focused primarily on the F1 score, but with the proliferation of large scale search engines, performance goals changed to place more emphasis on either precision or recall and so is seen in wide application.
The F-score is also used in machine learning. Note, however, that the F-measures do not take the true negative rate into account, and that measures such as the Matthews correlation coefficient, Informedness or Cohen's kappa may be preferable to assess the performance of a binary classifier.
The F-score has been widely used in the natural language processing literature, such as the evaluation of named entity recognition, Chinese word segmentation, etc. F-score is usually measured by IV F-score and OOV F-score, where the IV means in vocabulary and OOV means out of vocabulary. IV and OOV are distinguished by whether the testing words exist in the training data.
- Precision and recall
- NIST (metric)
- ROUGE (metric)
- Word Error Rate (WER)
- Receiver operating characteristic
- Matthews correlation coefficient
- Van Rijsbergen, C. J. (1979). Information Retrieval (2nd ed.). Butterworth.
- Beitzel., Steven M. (2006). On Understanding and Classifying Web Queries (Ph.D. thesis). IIT. CiteSeerX: 10.1.1.127.634.
- X. Li, Y.-Y. Wang, and A. Acero (July 2008). "Learning query intent from regularized click graphs". Proceedings of the 31st SIGIR Conference.
- See, e.g., the evaluation of the CoNLL 2002 shared task.
- Powers, David M W (2007/2011). "Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness & Correlation". Journal of Machine Learning Technologies 2 (1): 37–63.
- Aaron L.-F. Han, Derek F. Wong, and Lidia S. Chao (June 2013). "Chinese Named Entity Recognition with Conditional Random Fields in the Light of Chinese Characteristics". Proceedings of the 20th IIS Conference. LNCS Vol. 7912, pp. 57–68. Springer-Verlag Berlin Heidelberg.
- Aaron L.-F. Han, Derek F. Wong, Lidia S. Chao, Liangye He, Ling Zhu, and Shuo Li (September 2013). "A Study of Chinese Word Segmentation Based on the Characteristics of Chinese". Proceedings of the 25th GSCL Conference. LNCS Vol. 8105, pp. 111–118. Springer-Verlag Berlin Heidelberg.
|This statistics-related article is a stub. You can help Wikipedia by expanding it.|