caa a22 4500 46325144X CHVBK 20180405153341.0 cr unu---uuuuu 170326e20071201xx s 000 0 eng 10.1007/s10791-007-9034-8 doi (NATIONALLICENCE)springer-10.1007/s10791-007-9034-8 Diaz Fernando Department of Computer Science, University of Massachusetts-Amherst, 140 Governor's Drive, 01003-4610, Amherst, MA, USA aut Regularizing query-based retrieval scores [Elektronische Daten] [Fernando Diaz] We adapt the cluster hypothesis for score-based information retrieval by claiming that closely related documents should have similar scores. Given a retrieval from an arbitrary system, we describe an algorithm which directly optimizes this objective by adjusting retrieval scores so that topically related documents receive similar scores. We refer to this process as score regularization. Because score regularization operates on retrieval scores, regardless of their origin, we can apply the technique to arbitrary initial retrieval rankings. Document rankings derived from regularized scores, when compared to rankings derived from un-regularized scores, consistently and significantly result in improved performance given a variety of baseline retrieval algorithms. We also present several proofs demonstrating that regularization generalizes methods such as pseudo-relevance feedback, document expansion, and cluster-based retrieval. Because of these strong empirical and theoretical results, we argue for the adoption of score regularization as general design principle or post-processing step for information retrieval systems. Springer Science+Business Media, LLC, 2007 Regularization nationallicence Cluster hypothesis nationallicence Cluster-based retrieval nationallicence Pseudo-relevance feedback nationallicence Query expansion nationallicence Document expansion nationallicence Information Retrieval Springer Netherlands 10/6(2007-12-01), 531-562 1386-4564 10:6<531 2007 10 10791 https://doi.org/10.1007/s10791-007-9034-8 text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1007/s10791-007-9034-8 text/html Onlinezugriff via DOI NATIONALLICENCE 100 1- Diaz Fernando Department of Computer Science, University of Massachusetts-Amherst, 140 Governor's Drive, 01003-4610, Amherst, MA, USA aut NATIONALLICENCE 773 0- Information Retrieval Springer Netherlands 10/6(2007-12-01), 531-562 1386-4564 10:6<531 2007 10 10791 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer