PERFORMANCE MEASURES FOR INFORMATION EXTRACTION

By John I. Makhoul, F. Kubala, R. Schwartz, Ralph M. Weischedel. 2007

654 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Probabilistic Extension of Precision, Recall, and F1 Score for More Thorough Evaluation of Classification Models (2020, background): “Many NER speciﬁc metrics like SER (Makhoul et al., 1999) consider the possibility of having slot insertions or deletions, making them more appropriate for evaluating NER.”
Dynamic knobs for responsive power-aware computing (2011, methodology): “F-measure is the harmonic mean of the precision and recall.”
Semantic NLP-Based Information Extraction from Construction Regulatory Documents for Automated Compliance Checking (2016, background): “Thus, F-measure is defined as a weighted combination (harmonic mean) of precision and recall (Makhoul et al. 1999) [Eq.”
The ESTER phase II evaluation campaign for the rich transcription of French broadcast news (2005, methodology): “The error measure used was the slot error rate (SER) [ 5 ].”
The ETAPE corpus for the evaluation of speech-based TV content processing in the French language (2012, methodology): “Performance will be measured using the slot error rate (SER) metric (Makhoul et al., 1999).”
Overview of BioNLP Shared Task 2013 (2013, methodology): “The network prediction submissions have been evaluated against the reference network using an original metric, the Slot Error Rate (Makhoul et al. , 1999) that is more adapted to graph comparison than the usual Recall, Precision and F-score measures.”