Capturing user reading behaviors for personalized document summarization

Hao Jiang, Songhua Xu, Francis C.M. Lau

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

We propose a new personalized document summarization method, which observes a user's reading behaviors, including user facial expressions, gaze positions, and reading durations, during his or her past reading activities to infer the user's personal reading preferences. Once a user's personal reading preferences are derived, our algorithm can then automatically generate document summarization in a personalized way. We compare the performance of our algorithm with that of a few peer algorithms and software packages. The result of our comparative study shows that our algorithm can produce superior personalized document summaries than those peer methods in that the automatic document summarization generated by our algorithm can better satisfy a user's personal preferences.

Original languageEnglish (US)
Title of host publicationProceedings of the 15th International Conference on Intelligent User Interfaces, IUI'11
Pages355-358
Number of pages4
DOIs
StatePublished - 2011
Externally publishedYes
Event2011 15th ACM International Conference on Intelligent User Interfaces, IUI'11 - Palo Alto, CA, United States
Duration: Feb 13 2011Feb 16 2011

Publication series

NameInternational Conference on Intelligent User Interfaces, Proceedings IUI

Other

Other2011 15th ACM International Conference on Intelligent User Interfaces, IUI'11
Country/TerritoryUnited States
CityPalo Alto, CA
Period2/13/112/16/11

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction

Keywords

  • Facial expressions
  • Gaze positions
  • Personal preferences
  • Personalized document summarization
  • Reading durations
  • Reading preference

Fingerprint

Dive into the research topics of 'Capturing user reading behaviors for personalized document summarization'. Together they form a unique fingerprint.

Cite this