Agreement among statistical significance tests for information retrieval evaluation at varying sample sizes.
Mark D. SmuckerJames AllanBen CarterettePublished in: SIGIR (2009)
Keyphrases
- significance tests
- sample size
- hypothesis tests
- information retrieval evaluation
- hypothesis testing
- model selection
- test collection
- confidence intervals
- rank correlation
- upper bound
- statistical tests
- statistical power
- hypothesis test
- variance reduction
- supervised learning
- worst case
- random sample
- active learning
- np hard
- special case