In initial laboratory studies, subsymbolic user behavior1 has shown promise as a source of information for social navigation. Scalable, unobtrusive methods are needed for acquiring data on subsymbolic user behavior in field studies or live systems. Current methods are not suitable for use outside the laboratory because they interfere with normal user behavior and environment. We present a method for unobtrusively collecting subsymbolic user behavior in web-based systems, and report results from a field study. Our method is unobtrusive in that it uses current web technologies, works on the vast majority of current browsers, requires minimal instrumentation of existing web-based systems, and requires no additional user effort. This unobtrusive data collection method paves the way for future research on using subsymbolic user behavior to improve social navigation.
[1]
Kristina Höök,et al.
A recipe based on-line food store
,
2000,
IUI '00.
[2]
Petri Nokelainen,et al.
Enhancing the sense of other learners in student-centred Web-based education
,
2002,
International Conference on Computers in Education, 2002. Proceedings..
[3]
Peter Brusilovsky,et al.
Social Navigation Support in E-Learning : What are the Real Footprints ?
,
2005
.
[4]
Jude W. Shavlik,et al.
Learning users' interests by unobtrusively observing their normal behavior
,
2000,
IUI '00.
[5]
Yoshinori Hijikata,et al.
Implicit user profiling for on demand relevance feedback
,
2004,
IUI '04.
[6]
Mark Claypool,et al.
Implicit interest indicators
,
2001,
IUI '01.