On Using Social Context to Model Information Retrieval and Collaboration in Scientific Research Community - Community-Built Databases: Research and Development - page 134

and query representations and document-query matching [ 1 , 2 ]. These retrieval
approaches considered as system-centered ones are basically founded on computing
the topical relevance of a document according to a topical query. In fact, topical
relevance estimation is expressed using the document-query matching via evidence
showing the extent to which these two entities share contents expressed using
keywords or concepts. Thus, these approaches are characterized as “one size fits
all” since they provide the same results depending on the query keywords, even
though these latter are expressed by different users with different intentions, inter-
ests, backgrounds, and, more generally, surrounding contexts. While being funda-
mental for the advances and present stage of information retrieval, these approaches
make information retrieval difficult and challenging from the cognitive side, par-
ticularly in large-scale and interactive environments. The main criticism is that, in
these approaches, retrieval ignores the influence of the user's context that moves the
information relevance from topical to situational or cognitive one. According to the
cognitive view [ 3 ] that emerges from user-centered retrieval approaches, cognitive
relevance is leveraged by both topical relevance viewed from the system side and
usefulness viewed from the human side. Previous works in the field of contextual
information retrieval tackled the problem of the user-centered retrieval by combin-
ing search technologies and knowledge about the query and the user context into a
single framework to provide the most appropriate answer for the user's information
needs [ 4 ]. Context refers particularly to user's background, preferences, interests,
and community.
In this chapter, we focus on user's community as the key contextual factor for
enhancing the conception of several central concepts in the information retrieval
process, such as information, information need, user, interaction, and relevance.
More specifically, we address a particular community, namely the scientific com-
munity which is well known to be highly connected regarding both information
production and consumption. Thus, we investigate the ways to model scientific
collaboration for achieving information access. Accordingly, the main objectives of
this chapter are the following:
l To review the concept of context in information retrieval and highlight the shift
from personal to social context.
l To discuss major information retrieval models within scientific information
communities.
l To present and experimentally validate our approach of social information
retrieval specifically for scientific communities.
In Sect. 6.2 , we provide a basic background of notions relating to personal and
social context in information retrieval. We then give in Sect. 6.3 an overview of
collaboration and social retrieval models specifically designed for scientific infor-
mation access. Section 6.4 highlights the chapter's contributions. In Sect. 6.5 ,we
detail and evaluate the effectiveness of our approach of a social model for literature
access. Section 6.6 concludes the chapter and gives insight into the remaining
challenging issues.