Okapi is an experimental information retrieval system, running at the Centre for Interactive Systems Research at City University. The following sections describe earlier Okapi-related research, and introduce some of the salient issues for the ENQUIRE project.
The software on which Okapi's search engine was based was developed between 1982 and 1984 as part of a project to create an on-line library catalogue at the Polytechnic of Central London, (now the University of Westminster). Design of the search engine drew upon theoretical work on probabilistic retrieval by Steve Robertson and Karen Sparck-Jones [1].
In 1989 Okapi's developer, Steve Walker, moved to City University with the aim of building a system which could be used as a tool for investigating interactive retrieval. The ``Okapi at City'' project resulted in a system (hereafter called the VT100 version) which would be usable with different databases and had extensive transaction logging facilities to aid research.
Key features of Okapi which have remained important since the first operational implementation in 1989 are:
Since 1989, work has continued to refine the underlying search engine and to create user front-ends for different purposes. Throughout, Okapi has been maintained as an experimental retrieval system for which the key goal is to investigate the searching behaviour of real users.
Several Okapi projects have concentrated on the ways in which a query can be modified to optimise the searching process. The two methods to consider are query enhancement and query expansion.