Computer Science – Information Retrieval
Scientific paper
2012-02-13
Computer Science
Information Retrieval
Scientific paper
Recall, the proportion of relevant documents retrieved, is an important measure of a effectiveness in information retrieval, particularly in the legal, patent, and medical domains. For large document sets, recall can be estimated by assessing a random sample of documents for relevance; but a measure of the reliability of this estimate is also required. In this article, we examine several methods of calculating confidence intervals on recall estimates. We find that the normal approximation in current use provides poor coverage in many circumstances, even when adjusted to correct its inappropriate symmetry. Analytic and Bayesian methods based on the ratio of binomials are generally more accurate, but perform poorly on small populations. The recommended method derives beta-binomial posteriors on retrieved and unretrieved yield, with fixed hyperparameters, and a Monte Carlo estimate of the posterior distribution of recall. We demonstrate that this method gives mean coverage at or near the nominal level for differing scenarios, while being balanced and stable. We offer advice on sampling design, including the allocation of assessments to the retrieved and unretrieved segments, and compare the proposed beta-binomial with the officially reported normal intervals for recent TREC Legal Track iterations.
No associations
LandOfFree
Approximate Recall Confidence Intervals does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Approximate Recall Confidence Intervals, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Approximate Recall Confidence Intervals will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-88342