Advertisement

Predicting Retrieval Success Based on Information Use for Writing Tasks

  • Pertti Vakkari
  • Michael Völske
  • Martin Potthast
  • Matthias Hagen
  • Benno Stein
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11057)

Abstract

This paper asks to what extent querying, clicking, and text editing behavior can predict the usefulness of the search results retrieved during essay writing. To render the usefulness of a search result directly observable for the first time in this context, we cast the writing task as “essay writing with text reuse,” where text reuse serves as usefulness indicator. Based on 150 essays written by 12 writers using a search engine to find sources for reuse, while their querying, clicking, reuse, and text editing activities were recorded, we build linear regression models for the two indicators (1) number of words reused from clicked search results, and (2) number of times text is pasted, covering 69% (90%) of the variation. The three best predictors from both models cover 91–95% of the explained variation. By demonstrating that straightforward models can predict retrieval success, our study constitutes a first step towards incorporating usefulness signals in retrieval personalization for general writing tasks.

References

  1. 1.
    Belkin, N., Cole, M., Liu, J.: A model for evaluating interactive information retrieval. In: SIGIR Workshop on the Future of IR Evaluation, 23 July 2009, Boston (2009)Google Scholar
  2. 2.
    Hersh, W.: Relevance and retrieval evaluation: perspectives from medicine. J. Am. Soc. Inf. Sci. 45(3), 201–206 (1994)CrossRefGoogle Scholar
  3. 3.
    Järvelin, K., et al.: Task-based information interaction evaluation: the viewpoint of program theory. ACM Trans. Inf. Syst. 33(1), 3:1–3:30 (2015)Google Scholar
  4. 4.
    Cooper, W.S.: On selecting a measure of retrieval effectiveness. JASIST 24(2), 87–100 (1973)CrossRefGoogle Scholar
  5. 5.
    Vakkari, P.: Task based information searching. ARIST 1, 413–464 (2003)Google Scholar
  6. 6.
    Yilmaz, E., Verma, M., Craswell, N., Radlinski, F., Bailey, P.: Relevance and effort: an analysis of document utility. In: Proceedings of the CIKM 2014, pp. 91–100. ACM (2014)Google Scholar
  7. 7.
    Kelly, D., Belkin, N.J.: Display time as implicit feedback: understanding task effects. In: Proceedings of the SIGIR 2004, pp. 377–384. ACM (2004)Google Scholar
  8. 8.
    Liu, J., Belkin, N.J.: Personalizing information retrieval for multi-session tasks: the roles of task stage and task type. In: Proceedings of the SIGIR 2010, pp. 26–33. ACM (2010)Google Scholar
  9. 9.
    Liu, C., Belkin, N., Cole, M.: Personalization of search results using interaction behaviors in search sessions. In: Proceedings of the SIGIR 2012, pp. 205–214. ACM (2012)Google Scholar
  10. 10.
    Mao, J., et al.: Understanding and predicting usefulness judgment in web search. In: Proceedings of the SIGIR 2017, pp. 1169–1172. ACM, New York (2017)Google Scholar
  11. 11.
    Serola, S., Vakkari, P.: The anticipated and assessed contribution of information types in references retrieved for preparing a research proposal. JASIST 56(4), 373–381 (2005)CrossRefGoogle Scholar
  12. 12.
    Ahn, J.W., Brusilovsky, P., He, D., Grady, J., Li, Q.: Personalized web exploration with task models. In: Proceedings of the WWW 2008, pp. 1–10. ACM, New York (2008)Google Scholar
  13. 13.
    He, D., et al.: An evaluation of adaptive filtering in the context of realistic task-based information exploration. IP & M 44(2), 511–533 (2008)Google Scholar
  14. 14.
    Sakai, T., Dou, Z.: Summaries, ranked retrieval and sessions: a unified framework for information access evaluation. In: Proceedings of the SIGIR 2013, pp. 473–482. ACM (2013)Google Scholar
  15. 15.
    Potthast, M., Hagen, M., Völske, M., Stein, B.: Crowdsourcing interaction logs to understand text reuse from the web. In: Proceedings of the ACL 2013, pp. 1212–1221. Association for Computational Linguistics, August 2013Google Scholar
  16. 16.
    Potthast, M., et al.: ChatNoir: a search engine for the ClueWeb09 corpus. In: Proceedings of the SIGIR 2012, p. 1004. ACM, August 2012Google Scholar
  17. 17.
    Hagen, M., Potthast, M., Stein, B.: Source retrieval for plagiarism detection from large web corpora: recent approaches. In: CLEF 2015 Evaluation Labs, CLEF and CEUR-WS.org, September 2015Google Scholar
  18. 18.
    Hagen, M., Potthast, M., Völske, M., Gomoll, J., Stein, B.: How writers search: analyzing the search and writing logs of non-fictional essays. In: Kelly, D., Capra, R., Belkin, N., Teevan, J., Vakkari, P. (eds.) Proceedings of the CHIIR 2016, pp. 193–202. ACM, March 2016Google Scholar
  19. 19.
    Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.: Multivariate Data Analysis. Prentice-Hall, New Jersey (2010)Google Scholar
  20. 20.
    Hassan, A., Jones, R., Klinkner, K.: Beyond DCG: user behavior as a predictor of a successful search. In: Proceedings of the WSDM 2010, pp. 221–230. ACM (2010)Google Scholar
  21. 21.
    Gwizdka, J.: Characterizing relevance with eye-tracking measures. In: Proceedings of the 5th Information Interaction in Context Symposium, pp. 58–67. ACM (2014)Google Scholar
  22. 22.
    Smucker, M., Jethani, C.: Time to judge relevance as an indicator of assessor error. In: Proceedings of the SIGIR 2012, pp. 1153–1154. ACM (2012)Google Scholar
  23. 23.
    Weigl, D.M., Page, K.R., Organisciak, P., Downie, J.S.: Information-seeking in large-scale digital libraries: strategies for scholarly workset creation. In: Proceedings of the JCDL 2017, pp. 1–4, June 2017Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.University of TampereTampereFinland
  2. 2.Bauhaus-Universität WeimarWeimarGermany
  3. 3.Leipzig UniversityLeipzigGermany
  4. 4.Martin-Luther-Universität Halle-WittenbergHalleGermany

Personalised recommendations