Digital Content Management

The challenge for Digital Content Management research within Next Generation Localisation is to enhance and combine key aspects of Adaptive Hypermedia and Information Retrieval research to provide techniques, technology and prototype systems, to implement advanced content retrieval, slicing and adaptive composition of multilingual digital content. The key benefit for Next Generation Localisation is the increased ability to leverage multilingual documents in the Enterprise Localisation and Personalised Localisation processes, and to provide much greater flexibility in accessing, navigating and repurposing multilingual corpora. Some of the underlying research areas include:

Multilingual Information Access (MLIA) is concerned with the use of search requests in one language to retrieve documents in one or more other languages. The key challenge of MLIA beyond standard IR is crossing the language barrier between requests and documents. The specific task of retrieval between two languages is generally referred to as cross-language information retrieval (CLIR). While early work in MLIA and CLIR concentrated on published text documents, more recent work has extended this to explore multimedia IR and web data sources (CLEF 2000-2007).

Adaptive Hypermedia (AH) and Adaptive Web systems research is concerned with the dynamic composition and personalisation of hypermedia (WWW) documents in order to provide more context sensitive retrieval and reuse of digital content. In order to remedy for the negative effects of the traditional ‘one size fits all' approach to digital content, AH and adaptive web systems have the ability to adapt the information delivered based on the goals, tasks, and context of individual users and groups of users. AH research has typically supported three adaptation effects, namely adaptive content selection, adaptive navigation support and adaptive presentation. The principal application areas of adaptive web systems are in information kiosk-style systems, educational systems and tourism, although they are now emerging within museum information systems, eCommerce applications and eHealth.

In the Digital Content Management track, we seek to enhance and combine key aspects of Adaptive Hypermedia (and adaptive web systems) and Information Retrieval research to provide advanced annotation, slicing, retrieval and composition of multilingual digital content drawn from our corporate document repositories as well as open corpus origins e.g. WWW documents. Such digital content management research will enhance the impact of language technologies in Next Generation Localisation for Enterprise Localisation and provide unique capabilities and opportunities for Personalised Localisation.

Building on the innovative technologies provided by the Centre's ILT research track to analyse, annotate and translate content, the key challenge of DCM is to provide a step change in content localisation by enhancing user queries based on user context information and feedback, automate the generation of metadata required for localised content composition and support dynamic composition of localised content, customised for the user's needs and context of use.

The key objectives of the DCM research are to:

• Dynamically adapt user queries to provide richer, more contextualised queries which improve retrieval relevance and accuracy;
• Enhance language technologies with the capability of automatically generating metadata and semantic (subject) domain models suitable for adaptive multilingual hypermedia;
• Leverage ILT for content analysis to generate content and subject metadata and inform content slicing;
• Enhance language technology to support adaptive, dynamic composition of multilingual responses for enhanced user experience, cognitive comprehension and application;
• Support the validation and evaluation of these enhanced language technologies through various contrasting scenarios in Next Generation Localisation (Enterprise Level Localisation and Personalised Localisation), involving rich multimedia digital content.