• Aucun résultat trouvé

Knowledge Sharing in Information Seeking and Retrieval Situations

N/A
N/A
Protected

Academic year: 2022

Partager "Knowledge Sharing in Information Seeking and Retrieval Situations"

Copied!
2
0
0

Texte intégral

(1)

Knowledge Sharing in Information Seeking and Retrieval Situations

Preben Hansen1 and Claus-Peter Klas2

1 SICS - Swedish Institute of Computer Science

2 DistanceUniversity Hagen

preben@sics.se, claus-peter.klas@fernuni-hagen.de

Abstract. Research on information behaviour and information seeking has traditionally focused on the human as an individual as an information seeker and user of information. In this demonstration paper we present theDaffodilsystem as collaborative tool to reflect the social aspect on collaborative information retrieval.

1 An Experimental System for Collaborative Information Seeking and Retrieval

Research on information behaviour and the performance of the information seek- ing (IS) and retrieval tasks and processes has often been viewed in isolation. Re- cent studies undertaken in the areas of both work and everyday life IS show that people commonly work in groups and solve problems together with other people.

Collaborative information seeking and retrieval (CIS) is such an activity (e.g. [2], [3]). The social and collective aspects of human knowledge and experiences of information behaviour involve both CIS processes, but also the relationship be- tween people and the sharing of other people’s knowledge and experiences.This may enhance not just a single persons situation, but rather also a group of peo- ple. CIS includes how to search, analyse, judge information as well as use and share that with other people. It also includes ways of developing competence and knowledge as well as experinces. CIS involves finding, aggregating and exchang- ing information and knowledge and this can vary from an d hoc situation to a more planned team member effort. It views these processes as taking place and being embedded in work and other kinds of everyday life practices. Collabora- tion related to information seeking and retrieval may include sharing the same need for information, search strategies and results and further processing of the retrieved information: interpretation, filtering, synthesis or archiving potentially useful information into group repositories. Understanding the mutually shaping relationship between work practices as well as technologies yields a better insight into information and collaboration practices and enables a better understanding of how to develop systems for supporting those practices.

We introduce the Daffodil-System as an experimental system for CIS in the field of higher education in the domain of computer science.Daffodilis a virtual digital library system targeted at strategic support of users during the

Proceedings of the 1st Workshop on Social Information Retrieval for Technology-Enhanced Learning & Exchange

67

(2)

information seeking and retrieval process ([1]). It provides basic and high-level search functions for exploring and managing digital library objects including meta-data annotations over a federation of heterogeneous digital libraries (DLs).

In order to support such CIBDaffodilprovides on the one hand basiccollabo- rative services and on the other handrecommendations. Currently three special collaborative services. Thechat-toolenables direct communication between different users and between users and a possible help-desk. Also information ob- jects can be shared with the chat-tool via Drag&Drop. The personal library (PLib) enables users to store their information objects (DLOs) in a structured way. Through group folders a shared knowledge store is provided, where all users belonging to the group, can structure and store all informations. Through aware- ness mechanisms all users are always informed about changes. All DLOs can be annotated, either for personal use, like a summary, or in the group context, for discussion threads and ratings of stored objects. The PLib provides the bases forrecommendations and collaborative filtering. Depending on the stored information, which can be almost all DLOs, the task of the recommendation service is to suggest other information objects to users or groups and to support the formation of new groups. Additionally we currently investigate the possibil- ity to retrieve the best adaptive search suggestions for all users based on the current situation of an individual user in the search process. The ranked sugges- tions, comprises sixteen different suggestions ranging from terminological hints (e.g. vary spelling) to suggestions for using different tools from the Daffodil toolbox (e.g. show co-author network), are presented to each individual user and relay on case-based reasoning techniques to find the most useful suggestions for a given situation by comparing it to a case base of previous situations and adapting the solution. The case base consists of implicit logged user behaviour and explicit given feedback on the success of the suggestion.

Conclusions Our approach sees human information behaviour and information seeking and retrieval processes as firmly embedded in work and other social prac- tices, and views work practice, information objects, and information technologies as intertwined and mutually shaping. Importantly, the technologies we develop to facilitate CIB, increase the CIB structure of the workplace, and indeed in- crease the CIB structure in everyday life. The implemented collaboration feature in Daffodilcan be used as starting point to further research and evaluations.

References

1. N. Fuhr, C.-P. Klas, A. Schaefer, and P. Mutschke. Daffodil: An integrated desktop for supporting high-level search activities in federated digital libraries. In ECDL 2002, pages 597–612, Heidelberg et al., 2002. Springer.

2. P. Hansen and K. J¨arvelin. Collaborative information retrieval in an information- intensive work domain. Information Processing and Management, 41(5):1101–1119, 2005.

3. S. Talja and P. Hansen. New directions in human information behavior.Information Science and Knowledge Management: Information Sharing, 8:113–134, 2006.

Proceedings of the 1st Workshop on Social Information Retrieval for Technology-Enhanced Learning & Exchange

68

Références

Documents relatifs

Given the three collection selection approaches, namely TRD-CS, CORI using collection scores clustering, and optimal selection (OPT), and the three merging strategies LMS, CORI and

From combining various result lists (see Table 3 or 4), we cannot always obtain better retrieval effectiveness, where compared to isolated runs and the differences with the best single

Applying our logistic model using both the rank and the document score as explanatory variables, the resulting mean average precision is clearly better than the round-robin

The generalization of this theory, in order to consider fuzzy subsets as objects, was given in [5–9, 16, 17] using L-fuzzy sets as images and structuring elements, which was

FCART allows the user to load structured and unstructured data (texts and various meta- information) from heterogeneous data sources, build data snapshots, compose

Font specific attributes, such as bold, italic and case can be used in knowledge mapping and information retrieval to encode addi- tional data in texts, lists and labels to

The system then aggregates this body of evidence to predict the veracity of the claim, showing the user pre- cisely which information is being used and the various sources of

ture scholarly test collections.. We therefore argue that one or more well-crafted test collections with scholarly docu- ments is needed in order to progress significantly and