AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
Our evaluation showed that this instantiation of mediated collaboration improved selected precision, selected recall, viewed precision, and the number of unique relevant documents found compared with naive merging of search results obtained independently by two searchers

Algorithmic mediation for collaborative exploratory search

SIGIR, pp.315-322, (2008)

Cited by: 221|Views149
EI

Abstract

We describe a new approach to information retrieval: algorithmic mediation for intentional, synchronous collaborative exploratory search. Using our system, two or more users with a common information need search together, simultaneously. The collaborative system provides tools, user interfaces and, most importantly, algorithmically-mediat...More

Code:

Data:

0
Introduction
  • Information seeking can be more effective as a collaboration than as a solitary activity: different people bring different perspectives, experiences, expertise, and vocabulary to the search process.
  • The term “collaboration” has been used to refer to synchronous, intentionally-collaborative information seeking behavior
  • Such systems range from multiple searchers working independently with shared user interface awareness [9] to multiple people sharing a single user interface and cooperatively formulating queries and evaluating results [13].
  • While awareness of one’s co-searcher(s) is an important first step for collaborative retrieval, user interface-only solutions still require too much attention to others’ results.
  • In these user interface-only systems, searchers must manually reconcile and integrate their activities with their co-searcher(s)
Highlights
  • Information seeking can be more effective as a collaboration than as a solitary activity: different people bring different perspectives, experiences, expertise, and vocabulary to the search process
  • Collaboration goes beyond the user interface: Information that one team member finds is not just presented to other members, but it is used by the underlying system in realtime to improve the effectiveness of all team members while allowing each to work at their own pace
  • The term “collaboration” has been used to refer to synchronous, intentionally-collaborative information seeking behavior. Such systems range from multiple searchers working independently with shared user interface awareness [9] to multiple people sharing a single user interface and cooperatively formulating queries and evaluating results [13]
  • Our evaluation showed that this instantiation of mediated collaboration improved selected precision, selected recall, viewed precision, and the number of unique relevant documents found compared with naive merging of search results obtained independently by two searchers
  • Different media types may require different user interfaces to elicit queries and to display results, the underlying mediation need not change because this particular mediation algorithm, supporting these roles, is content-domain independent
  • Numerous challenges remain, including designing and comparing different realtime merging strategies for query results, defining additional roles, better understanding the tradeoffs between parallel and synchronized work, and designing appropriate user interfaces
Methods
  • The authors used a mixed-design experimental method, where teams of searchers performed one of two search conditions; all 24 TRECvid interactive retrieval topics were used in both conditions.
  • Eight people participated in the experiment.
  • No participants had prior experience with the topics on which they searched.
  • In both conditions teams had 15 minutes to complete a topic; each condition consisted of two team members working for 15 minutes, for a total of 30 person-minutes per topic
Results
  • The authors wanted to test the hypothesis that mediated collaboration offers more effective searching than post hoc merging of independently produced results, as was done, for example by Baeza-Yates et al [4].

    5.2.1 Collaboration

    To assess the teams’ performance, the authors removed duplicate documents from the merged result set, and kept track of which documents each person saw.
  • The authors wanted to test the hypothesis that mediated collaboration offers more effective searching than post hoc merging of independently produced results, as was done, for example by Baeza-Yates et al [4].
  • Participants in the merged condition saw an average of 2978 distinct documents per topic; participants in the collaborative condition saw an average of 2614 distinct documents per topic.
  • At the end of the 15 minute session, Rs was 29.7% higher for collaborative search than for merged results.
Conclusion
  • Built, and evaluated a system that mediates search for a focused team of searchers.
  • The authors' evaluation showed that this instantiation of mediated collaboration improved selected precision, selected recall, viewed precision, and the number of unique relevant documents found compared with naive merging of search results obtained independently by two searchers.
  • Numerous challenges remain, including designing and comparing different realtime merging strategies for query results, defining additional roles, better understanding the tradeoffs between parallel and synchronized work, and designing appropriate user interfaces.
  • The authors are confident that these first steps will lead to a fruitful research field, success in which will rely on the combined efforts of IR and HCI researchers
Tables
  • Table1: Precision for single-query runs vs. precision for 2-way and 3-way fused runs. indicates t-test significance at p < 0.01
  • Table2: Average percent improvement of collaborative over merged, at various time points
  • Table3: Average ranks for uniqueness measures. Smaller scores represent better performance. The “%Chg” column represents the percent improvement (decrease in average rank) of collaborative over merged
Download tables as Excel
Reference
  • J. Adcock, M. D. Cooper, A. Girgensohn, and L. Wilcox. Interactive video search using multilevel indexing. In CIVR 2005, pages 205–214, 2005.
    Google ScholarLocate open access versionFindings
  • J. A. Aslam and M. Montague. Models for metasearch. In W. B. Croft, D. J. Harper, D. H. Kraft, and J. Zobel, editors, Proc. SIGIR 2001, pages 276–284. ACM Press, September 2001.
    Google ScholarLocate open access versionFindings
  • J. A. Aslam, V. Pavlu, and R. Savell. A unified model for metasearch and the efficient evaluation of retrieval systems via the hedge algorithm. In Proc. SIGIR 2003, pages 393–394, July 2003.
    Google ScholarLocate open access versionFindings
  • R. Baeza-Yates and J. A. Pino. A first step to formally evaluate collaborative work. In GROUP ’97: Proc. ACM SIGGROUP Conference on Supporting Group Work, pages 56–60, New York, NY, USA, 1997.
    Google ScholarLocate open access versionFindings
  • N. J. Belkin, P. Kantor, E. A. Fox, and J. A. Shaw. Combining the evidence of multiple query representations for information retrieval. Information Processing and Management, 31(3):431–448, 1995.
    Google ScholarLocate open access versionFindings
  • A. Girgensohn, J. Adcock, M. D. Cooper, and L. Wilcox. A synergistic approach to efficient interactive video retrieval. In INTERACT, pages 781–794, 2005.
    Google ScholarLocate open access versionFindings
  • G. Golovchinsky. What the query told the link: the integration of hypertext and information retrieval. In HYPERTEXT 1997, pages 67–74, New York, NY, USA, 199ACM.
    Google ScholarLocate open access versionFindings
  • A. Hauptmann, W.-H. Lin, R. Yan, J. Yang, and M.-Y. Chen. Extreme video retrieval: Joint maximization of human and computer performance. In Proc. ACM Multimedia 2006, pages 385–394, Santa Barbara, CA, 2006.
    Google ScholarLocate open access versionFindings
  • M. R. Morris and E. Horvitz. Searchtogether: an interface for collaborative web search. In Proceedings of UIST, pages 3–12, 2007.
    Google ScholarLocate open access versionFindings
  • J. Pickens and G. Golovchinsky. Collaborative exploratory search. In Proc. 2007 HCIR Workshop, pages 21–22, October 2007.
    Google ScholarLocate open access versionFindings
  • C. Shah, D. Kelly, and X. Fu. Making mind and machine meet: a study of combining cognitive and algorithmic relevance feedback. In Proc. SIGIR 2007, pages 877–878, New York, NY, USA, 2007. ACM.
    Google ScholarFindings
  • J. A. Shaw and E. A. Fox. Combination of multiple searches. In Text (REtrieval) Conference, pages 105–108, 1994.
    Google ScholarLocate open access versionFindings
  • A. F. Smeaton, H. Lee, C. Foley, S. McGivney, and C. Gurrin. Fıschlar-diamondtouch: Collaborative video searching on a table. In Multimedia Content Analysis, Management, and Retrieval, San Jose, CA, January 15-19 2006.
    Google ScholarLocate open access versionFindings
  • B. Smyth, E. Balfe, O. Boydell, K. Bradley, P. Briggs, M. Coyle, and J. Freyne. A live-user evaluation of collaborative web search. In Proc. IJCAI 2005), pages 1419–1424, Ediburgh, Scotland, 2005.
    Google ScholarLocate open access versionFindings
  • M. B. Twidale, D. M. Nichols, and C. D. Paice. Browsing is a collaborative process. Information Processing and Management, 33(6):761–783, 1997.
    Google ScholarLocate open access versionFindings
Your rating :
0

 

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科