DOI: 10.7763/IPEDR. 2012. V51. 1 CULTURA: Supporting Enhanced Exploration of Cultural Archives through Personalisation Eoin Bailey 1 + , Seamus Lawless 1 , Cormac Hampson 1 , Alexander O’Connor 1 , Mark Sweetnam 2 , Owen Conlan 1 and Vincent Wade 1 1 Knowledge and Data Engineering Group, School of Computer Science and Statistics, Trinity College Dublin, Ireland 2 School of History and Humanities, Trinity College Dublin, Ireland Abstract. A key challenge facing curators and providers of digital cultural heritage across Europe and Worldwide is to instigate, increase and enhance engagement with cultural archives. To achieve this, a fundamental change in the way cultural resources are experienced and contributed to by communities is required. Furthermore, it is of central importance to create systems which can support the full spectrum of the user community - from professional researchers engaged in answering complex queries, to supporting novice students or members of the public in finding their way through the vast collection of resources. This paper presents CULTURA, a personalisation environment for navigating digitised cultural heritage archives. An initial description of a case study surrounding the 1641 Depositions Archive is followed by an analysis of the results of trials with students in the humanities. Keywords: Personalisation, Digital Cultural Heritage, CULTURA, User Modelling. 1. Introduction Substantial effort has been expended in the digitisation and preservation of cultural heritage collections. Traditionally, this effort has been focused on the creation of digital representations of cultural artefacts, and the creation of metadata and documentation associated with these artefacts. As a result, a continually expanding volume of content is now available to humanities scholars, in a variety of formats. After digitisation, these collections are typically monolithic and can often be difficult to search and navigate. Historic manuscript collections often contain text which is highly inconsistent in terms of language, spelling, punctuation and terminology as language and spelling were rarely standardised. Existing platforms [1-3] for digital archives tend either to provide very basic tools for dealing with a range of collections, or more complex tools whose utility is tied to a very specific type of collection, or even to a single archive [4]. By contrast, CULTURA is designed to be a corpus-agnostic platform that provides tools for the detailed exploration and interrogation of a range of digital collections. For this reason, its development to date has involved not close cooperation between humanities scholars and computer scientists. CULTURA is being developed around two very different content collections. Trinity College Dublin is providing the 1641 Depositions [5], a textual source, and the University of Padua is providing the contents of the Imaginum Patavinae Scientiae Archivum (IPSA) [6], an image-based collection. 1.1. Motivation Understanding a large collection of documents presents a significant challenge, a global and micro view of the documents is required and views will also vary depending on the experience and knowledge of the user. Additionally the language in the documents themselves poses a challenge, for example variations in spelling of words, or the change of language over time. The digitisation process opens these documents up to an array of techniques and processes to aid in these issues, while also making the fragile documents available to a much wider audience. Additionally, different tiers of researchers and users have different tasks, goals, and scaffolding around their research. An environment that supports professional discoveries and new + Corresponding author. Tel.: + 353 1 896 8431; fax: + 353 1 677 2204. E-mail address: eoin.bailey@scss.tcd.ie. 1