Empirical evidence for search effectiveness models

Alfan Farizki Wicaksono, Alistair Moffat

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

17 Citations (Scopus)

Abstract

Given a SERP in response to a user-originated query, Moffat et al. (CIKM 2013; TOIS 2017) suggest that C(i), the conditional continuation probability of the user examining the (i + 1) st element presented in the SERP, given that they are known to have examined the i th one, is positively correlated with both i and with the user's initial estimate of the volume of answer pages they are looking for, and negatively correlated with the extent to which suitable answer pages have been identified in the SERP at positions 1 through i. Here we first describe a methodology for specifying how C(i) should be defined in practical (as against ideal) settings, and then evaluate the applicability of the approach using three large search interaction logs from two different sources.

Original languageEnglish
Title of host publicationCIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management
EditorsNorman Paton, Selcuk Candan, Haixun Wang, James Allan, Rakesh Agrawal, Alexandros Labrinidis, Alfredo Cuzzocrea, Mohammed Zaki, Divesh Srivastava, Andrei Broder, Assaf Schuster
PublisherAssociation for Computing Machinery
Pages1571-1574
Number of pages4
ISBN (Electronic)9781450360142
DOIs
Publication statusPublished - 17 Oct 2018
Event27th ACM International Conference on Information and Knowledge Management, CIKM 2018 - Torino, Italy
Duration: 22 Oct 201826 Oct 2018

Publication series

NameInternational Conference on Information and Knowledge Management, Proceedings

Conference

Conference27th ACM International Conference on Information and Knowledge Management, CIKM 2018
Country/TerritoryItaly
CityTorino
Period22/10/1826/10/18

Keywords

  • Adaptive metric
  • Average precision
  • Evaluation
  • User model

Fingerprint

Dive into the research topics of 'Empirical evidence for search effectiveness models'. Together they form a unique fingerprint.

Cite this