A. Bookstein. When the most "pertinent" document should not be retrieved { an analysis of the swets model. Information Processing and Management, 13:377-383, 1977. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Applying Semantic Relation Extraction to Information Retrieval - Loper (2000) (1 citation) (Correct)
....vectors take, and on the time it takes to compare them [26] 2. 2 Measuring Information Retrieval Performance Since the early 1960 s, much research has gone into nding good metrics with which to measure information retrieval performance [24] Since then, many alternative metrics have been proposed[24, 7, 21, 3]. However, only a few have become widely accepted. Most of the widely accepted metrics are based on the contingency table, which divides the sentences of the original corpus into four categories, based on whether they answer the question and on whether they are retrieved in response to the ....
A. Bookstein. When the most "pertinent" document should not be retrieved { an analysis of the swets model. Information Processing and Management, 13:377-383, 1977.
Modeling Score Distributions for Combining the Outputs of.. - Manmatha, Rath, Feng (2001) (17 citations) (Correct)
....the relevant data with a Gaussian. A number of researchers in the 60 s and 70 s starting with Swets [17] proposed fitting both the relevant and non relevant scores using normal distributions and then using statistical decision theory to find a threshold for deciding what was relevant. Bookstein [3] pointed out that Swets implicitly relied on an equal variance assumption. Bookstein also raised the issue of whether it might be more appropriate to model the score distributions using Poissons. This modeling does not appear to have been done. van Rijsbergen [19] commented that for search engines ....
A. Bookstein. When the most "pertinet" document should not be retrieved - an analysis of the swets model. Information Processing and Management, 13:377--383, 1977.
Relevance: The Whole (hi)story - Mizzaro (1996) (29 citations) (Correct)
....T 1972 Foskett [Fos72] C 1973 Belzer [Bel73] EC Cooper [Coo73b] C Cooper [Coo73a] C Thompson [Tho73] E Wilson [Wil73] C 1974 Kemp [Kem74] C Kochen [Koc74] T 1975 Saracevic [Sar75] S 1976 Saracevic [Sar76] S Table 1: Papers about relevance Part II Year Author(s) Ref. Type 1977 Bookstein [Boo77] T Davidson [Dav77] E Maron [Mar77] C Robertson [Rob77] SC Swanson [Swa77] C Tessier, Crouch, Atherton [TCA77] C 1978 Cooper, Maron [CM78] T Figueiredo [Fig78] E Marcus, Kugel, Benenfeld [MKB78] E Wilson [Wil78] C 1979 Bookstein [Boo79] CT Kazhdan [Kaz79] E Koll [Kol79] E Lancaster [Lan79] CS 1980 ....
....Observing this line of research in the two periods, it is easy to note the growing presence of user oriented, cognitive approaches, in place of purely mathematical ones. 6. 2 Kinds The analysis of various kinds of relevance continues, often in a short sighted way (see Section 2) Bookstein [Boo77, Boo79] distinguishes between the relevance of a document, assigned by the user, and the prediction of relevance of a document, assigned by the IRS and called Retrieval Status Value (RSV) He notes that RSV and relevance might not coincide, and so the documents with the highest RSVs might not be ....
A. Bookstein. When the most "pertinent" document should not be retrieved-- An analysis of the Swets model. Information Processing & Management, 13(6):377--383, 1977.
Information Retrieval - van Rijsbergen (1979) (411 citations) (Correct)
No context found.
BOOKSTEIN, A., 'When the most "pertinent" document should not be retrieved - an analysis of the Swets Model', Information Processing and Management, 13, 377-383 (1977).
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC