Information Technology Reference
In-Depth Information
lowing this rule precludes any exploration, and is
sub-optimal. Indeed, initially, a priori probability
that any message should be transmitted is too low
to justify transmitting it. Therefore there must
be some period of exploratory transmission to
establish a rule. Heuristics of the form: “explore
until some criterion C is met, and do not explore
thereafter” have, with various escape clauses,
produced the best performance in the TREC2004
evaluation (CAS 2004). The effect of variation in
the learning model itself is given by (Fradkin and
Kantor, 2004, 2005). Models that concentrate on
the conditional posterior probabilities that a mes-
sage should be transmitted have been explored by
(Elovici, Shapira, Kantor; 2003, 2005).
proposed categories; second, improve the learn-
ing models for automatically access information
objects against those categories; last, design and
conduct evaluation of the impact of the system
on the work product of user groups.
reFerenCes
Argamon, S., Koppel, M., Fine, J., & Shimoni, A.
R. (2003). Gender, Genre, and Writing Style in
Formal Written Texts. Text , 23 (3). doi:10.1515/
text.2003.014
Bai, B., Ng, K. B., Sun, Y., Kantor, P., &
Strzalkowski, T. (2004). The institutional di-
mension of document quality judgments. In
the Proceedings of the 2004 Annual Meeting of
American Society for Information Science and
Technology .
ConClusion
Collaborative search, or social search, has been
predicted to be the next big thing on the Internet.
Both Yahoo! and Microsoft have released their
collaborating search tools. It is easy to see that
there are many issues involved in the collabora-
tive searching system design. The focus of the
technology development, as well as research, is
mainly on the design of interface to facilitate the
collaboration of searching. In this chapter we lim-
ited our discussion from a different perspective:
quest representation and matching. We proposed
the idea of construction of a richer query profile
(Quest) above and beyond the topicality feature of
the user's information need which is currently in
common use. Quests will be shared by searchers
and will be simply picked up by a future user if
the Quest is similar enough to his/her information
need context.
To make the proposed idea practical, much
research is needed to, first, identify a set of
characterizes that are important to be involved in
Quest (query profile), we proposed two sets such
properties in this chapter based on previous works,
a study with current web users, especially social
network frequent users will be helpful to update the
Barouni-Ebrahimi, M., & Ghorbani,A.A. (2008).
An interactive search assistant architecture
based on intrinsic query stream characteristics.
Computational Intelligence , 24 (2), 158-190.
doi:10.1111/j.1467-8640.2008.00326.x
Barry, C. L. (1994). User-defined relevance
criteria: An exploratory study. Journal of the
American Society for Information Science
American Society for Information Science ,
45 (3), 149-159. doi:10.1002/(SICI)1097-
4571(199404)45:3<149::AID-ASI5>3.0.CO;2-J
Bruce, H. W. (1994). A cognitive view of the
situational dynamism of user-centered relevance
estimation. Journal of the American Society for
Information Science American Society for Infor-
mation Science , 45 (3), 142-148. doi:10.1002/
(SICI)1097-4571(199404)45:3<142::AID-
ASI4>3.0.CO;2-6
Chaski, C. (2001). Empirical evaluations of
language-basedAuthor identification techniques.
Forensic Linguistics . The International Journal
of Speech Language and the Law , 8 (1).
Search WWH ::




Custom Search