JISC

Content deposited in institutional repositories is growing, however there is no comprehensive and easy way to search and retrieve this content. We believe that, in order to facilitate access to scholarly and educational material, this content should be widely accessible to the UK education community. In partnership with UKOLN and SHERPA, Intute has been commissioned by JISC to develop a repository search infrastructure. This development will build on the ePrints UK project, and aims to facilitate the discovery, access and retrieval of material.

The project is in 3 phases. 

• Extending the scalability of the search technology and infrastructure
• Development of search and aggregation services allowing metadata enhancement
• Development of semantic and contextual search and discovery features.

In doing so, we hope to raise the visibility of repository content and perpetuate the deposit of content.

Overview

Within the UK University context there are a large number of digital information repositories being used and potentially of use to the community.  In many areas these are subject-specific and are, to some extent, in the nature of inaccessible silos.

The key aim of this project is to develop a search and discovery middle layer that can conduct cross-repository search and discovery by context, expose, aggregate and de-duplicate information artefacts and make them available to a user search or drive a personalized information space.

This project builds on extensive work into academic-oriented user requirements relating to search and discovery.  What has become clear early on is that simple ‘Google-type’ search is not sufficient.  Rather, there is a need to build in scalable contextual and discovery oriented features that address the complexities as illustrated below.

Aims and objectives 

• Continually extend the coverage, scope and functionality of the search and discovery service
• Extend capacity to enhance and re-export metadata
• Generate cultural buy-in and embedding to deposit in Open Access repositories by facilitating access through search services
• Develop a thorough understanding of stakeholder needs (especially end users and repository managers) in order to shape project development
• Explore synergies / alliances with a range of JISC services and projects such as UKOLN, AHDS, MIMAS, Edina and SHERPA, as well as with other projects within the JISC Repositories and Preservation programme
• Explore international collaboration with repository search services
• Facilitate ubiquitous access (including machine to machine) to the UK Repository Search through a variety of first points of contact e.g. Web 2.0 personal spaces, VREs, VLEs, institutional websites, portals, catalogues, OpenURL routers etc
• Design and model the UK Repository Search both as a comprehensive tool for the UK and a model from a global perspective

Anticipated outputs and outcomes

• cross repository discovery, aggregation and retrieval of information, from all HE and other relevant repositories across the UK by author, title and subject.  The Intute Repository Search (IRS) will be a primary aggregator adding quality
• exploration of issues in searching across full-text
• exploitation of synergies between research oriented repositories and learning object repositories, both at the institutional, subject and national level (e.g. JORUM)
• ability to personalize information based on a user profile, task-based, context or priority
• leveraging our metadata investments to allow aggregation, concept or high-value retrieval and automatic or guided development of taxonomies
• identification of individuals by interest and areas of expertise thus building knowledge bridges both within and across institutions and subject domains
• identifying and developing high-value research knowledge communities
• embedding repositories in the familiar and day-to-day Research Desktop environment
• enable a target/estimation of the amount of content being cross-searched and contribute to the development of  metrics supporting the research appraisal process
• 'showcase’ the collective and collaborative UK research output and perspectives

Technology / Standards used (if applicable)

Name of standard or specification	Version
OAI - PMH	V2
Dublin Core	DCAP
JSR168	V1
UIMA
SRU / SRW
RSS	1.0 / 1.2
WSRP

Lead institution

• Intute (MIMAS, University of Manchester)

Project partners

• UKOLN University of Bath
• SHERPA University of Nottingham
• NaCTeM (National Centre for Text Mining) University of Manchester