User model of information retrieval on the Internet: a cognitive psychological approach
K.W. Sandberg, Y. Pan and J. Palme
Department of Informatics, Mid Sweden University, Frösön, Sweden
Information retrieval has changed considerably in the last years with the expansion of the Web (World Wide Web) and the advent of modern and inexpensive graphical user interfaces and mass storage devices. How does one find anything on the Internet? There are essentially four methods of information retrieval on Internet: browsing, consulting an index, word of mouth, or posting a request.
Problem in information
retrieval
The key problems in Internet information retrieval, in particular concerned
with ill-defined information requirements, concept interpretation in searching
and text representation in indexing. The main objective of information retrieval
is to facilitate the effective communication of desired information between
a human generator of information and the human user. This communication takes
place in an information system.
Aims
The present paper aims to analyse and suggest solutions to key problems in online
information retrieval, in particular concerned with the search of information
problems on Internet.
User model of information
retrieval
Information retrieval research has developed in a fairly complex manner around
the classic but effective vector space model, abstracting from single keywords
to cluster of keywords and other increasingly abstract objects, in order to
improve discriminating performance. User model from the psychological point
of view represent the category of keywords with associated score according to
the "mental representation" present in the model. We use conceptual graphs to
model the user's search goal on Internet. The nodes represent keywords, annotated
with frequency of usage; the links represent the distance between keywords in
the model. The task isto transform user requests into queries by understanding
the information problem or the underlying goal or problem.
Design and usability
testing
Efficient usability testing involves choosing the most appropriate methods for
the problem at hand. The purpose of testing is to provide design support and
the emphasis should therefore be on obtaining the maximum design-related information
for the testing effort. In the present discussion the research is reserved for
that more general activity that is directed towards the accumulation of general
principles. When specific answers to very specific questions are required the
usability testing is more appropriate.
Poster presented at Measuring Behavior 2000, 3rd International Conference on Methods and Techniques in Behavioral Research, 15-18 August 2000, Nijmegen, The Netherlands
© 2000 Noldus Information Technology b.v.