Text Access Potentials for Interactive Information Retrieval

TAPIR started up as a research project in June 2001. In 2002 the project is sponsored by NORDINFO and the Research Council of The Danish Ministry of Culture. TAPIR aims at investigating the potentials of applying the diversity of cognitive representations pointing to scientific full-text documents following the principle of poly-representation. Poly-representation (or multi evidence) implies to utilize the cognitively different overlapping interpretations, also over time, made by different actors participating in interactive IR. Such cognitive overlaps derive, for instance, from the authors´ own perceptions of their work (titles, full-text terms), from human indexing (e.g. descriptors), or from citations given to the work by other authors. The assumption is that the more cognitively different the representations simultaneously pointing to a document are, the higher is the probability that the document is relevant to a given set of criteria. The theoretical framework is constituted by the cognitive approach to IIR as developed by Ingwersen (1992; 1996), Ingwersen and Christensen (1997), and Borlund (1997; 1998; 2000a, 2000b).

With the aim in mind, TAPIR seeks to build up a number of scientific domain-related full-text test collections. TAPIR applies the IIR Evaluation Package consisting of simulated work task situations and a set of procedures for carrying out IIR evaluations in dedicated information environments. The simulated situations are applied as triggers for domain test persons´ interpretations and information need developments as well as acting as platforms for human relevance assessments during evaluation of IR performance and information seeking behaviour. Relevance dimensionality, criteria, and scaling issues are central to the TAPIR project which can be seen as a modest follow up of the OKAPI project.

Thus far, TAPIR attempts to build up and test poly-representative algorithms involving scientific references and the citation networks in specific collections within Medicine or Engineering. This part of the project is basically sponsored by a grant (# A2001 06-021 (a), 504.000 DKK) from the Ministry of Culture. The TAPIR Team also conducts the Web-TAPIR line of research, in parallel with TAPIR. Web-TAPIR is sponsored by a grant (25.000 Euro) from NORDINFO, the Nordic Council for Scientific Information.

Web-TAPIR aims at the exploration of the diversity of cognitive representations pointing to scientific Web documents, also following the principle of poly-representation. In order to develop and test such algorithms, that also include the application of metadata, Web-page contents and link structures, etc., dedicated Web collections are under development.