CLSep 5, 2018
Appendix - Recommended Statistical Significance Tests for NLP Tasks
arXiv:1809.01448v11 citations
Originality Synthesis-oriented
AI Analysis
This work provides incremental guidance for NLP researchers to ensure robust experimental conclusions.
The paper addresses the need for valid statistical significance tests in NLP by proposing specific tests for common tasks and evaluation measures, complementing an existing guide.
Statistical significance testing plays an important role when drawing conclusions from experimental results in NLP papers. Particularly, it is a valuable tool when one would like to establish the superiority of one algorithm over another. This appendix complements the guide for testing statistical significance in NLP presented in \cite{dror2018hitchhiker} by proposing valid statistical tests for the common tasks and evaluation measures in the field.