No abstract

Eero Sormunen. 2004. Query exhaustivity, relevance feedback and search success in automatic and interactive query expansion. Journal of Documentation 60:2, 109-127. [Abstract] [Full Text] [PDF]26. Jane Greenberg.

No abstract

The paper aims to explore the concepts of interaction and interactivity presented in different theoretical models in the fields of human-computer interaction (HCI) and information-seeking/searching behaviour, and to relate these to information retrieval (IR) research. It is suggested that interaction in HCI is primarily concerned with establishing a user/system dialogue at the user interface and does not address the interactive characteristics of IR operational tasks. A distinction is made between general informationseeking models and information-searching models for computerised systems. The former are deemed to provide a useful framework for characterising interaction at the task level, with the structural relationship between tasks as well as the dynamic transition from one task to another being key features of the interactive process. Although the latter are all concerned with how searchers interact with IR systems, each of the models examined represents user interaction at different levels of abstraction. Taken together they provide complementary views of a highly dynamic process. Three principal aspects of interaction are identified and discussed: interaction within and across tasks; the notion of interaction as task sharing; and interaction as a discourse. In conclusion the adoption of an interaction paradigm for IR research is advocated and examples of empirical work for supporting interactive searching and retrieval are provided. 431 431

A brief review of the history of laboratory testing of information retrieval systems focuses on the idea of a general‐purpose test collection of documents, queries and relevance judgements. The TREC programme is introduced in this context, and an overview is given of the methods used in TREC. The Okapi team’s participation in TREC is then discussed. The team has made use of TREC to improve some of the automatic techniques used in Okapi, specifically the term weighting function and the algorithms for term selection for query expansion. The consequence of this process has been a very good showing for Okapi in terms of the TREC evaluation results. Some of the issues around the much more difficult problem of interactive evaluation in TREC are then discussed. Although some interesting interactive experiments have been performed at TREC, the problems of reconciling the requirements of the laboratory context with the concerns of interactive retrieval are still largely unresolved.