TY - GEN
T1 - On theoretically valid score distributions in information retrieval
AU - Cummins, Ronan
AU - O'Riordan, Colm
PY - 2012
Y1 - 2012
N2 - In this paper, we aim to investigate the practical usefulness of the Recall-Fallout Convexity Hypothesis (RFCH) for a number of document score distribution (SD) models. We compare SD models that do not automatically adhere to the RFCH to modified versions of the same SD models that do adhere to the RFCH. We compare these models using the inference of average precision as a measure of utility. For the three models studied in this paper, we conclude that adhering to the RFCH is practically useful for the two-normal model, makes no difference for the two-gamma model, and degrades the performance of the two-lognormal model.
AB - In this paper, we aim to investigate the practical usefulness of the Recall-Fallout Convexity Hypothesis (RFCH) for a number of document score distribution (SD) models. We compare SD models that do not automatically adhere to the RFCH to modified versions of the same SD models that do adhere to the RFCH. We compare these models using the inference of average precision as a measure of utility. For the three models studied in this paper, we conclude that adhering to the RFCH is practically useful for the two-normal model, makes no difference for the two-gamma model, and degrades the performance of the two-lognormal model.
UR - https://www.scopus.com/pages/publications/84860147748
U2 - 10.1007/978-3-642-28997-2_39
DO - 10.1007/978-3-642-28997-2_39
M3 - Conference Publication
AN - SCOPUS:84860147748
SN - 9783642289965
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 451
EP - 454
BT - Advances in Information Retrieval - 34th European Conference on IR Research, ECIR 2012, Proceedings
T2 - 34th European Conference on Information Retrieval, ECIR 2012
Y2 - 1 April 2012 through 5 April 2012
ER -