Digital Content Management

Goals

The key challenge of the DCM research track is to provide a step change in content localisation focusing on three areas: user query enhancement, metadata and model development, and dynamic composition of localised content, customised for the users’ needs and context of use. The DCM track is divided between these three work areas called DCM 1, DCM 2 and DCM 3:

• Enhancement of user queries based on user context information and feedback
  (DCM1)
• Automation and semi-automation of the models and generation of metadata
  required for localised content composition (DCM 2)
• Support for dynamic composition of localised content, customised for the user’s need and context (DCM 3)

However, the research is integrated across the work areas via combined prototypes
and experiments. Key prototypes are integrated across research tracks (i.e. ILT, LOC, SF) via demonstrator systems. Principally the DCM prototypes are being used within the Personalised Multilingual Customer Care Management Demonstrator (PMCC) and will be used within the Personalised Multilingual Social Networking
Demonstrator (PMSN).

Methodology

With the increasing volume of digital content and the diversity of devices upon which localised content needs to be rendered, it is becoming impossible to manually annotate, slice and compose appropriate localised content. In addition, localisation needs to not only adapt to suit specific corporate localisation requirements, but also satisfy individual user needs for localised translations. The three principal areas of DCM research relate to the challenges of locating and retrieving content; of modelling knowledge in a structured, reusable way; and in supporting the user by harnessing adaptivity to give users significantly improved access to the information they need. A central theme running through all of these challenges is the need to provide the information in a form that is tailored to the user’s requirements and preferences, and which includes not only the direct response to their query, but key supporting information that the user might need to achieve their goal. The challenge for DCM within Next Generation Localisation is to enhance and combine key aspects of Adaptive Hypermedia (AH) and Information Retrieval (IR) research to provide techniques, technology and prototype systems, to implement advanced content retrieval, slicing and adaptive composition of multilingual digital content. The DCM1 work package addresses the issues of IR research directly. There is a large community of research involved in IR, particularly on web data. DCM1 research includes the application of cross-lingual techniques to permit users to gain access to information not in their native tongues. Personalisation in IR is addressed both in the use of user modelling techniques to alter the behaviour of IR systems, and also through the creation of hybrid Adaptive IR systems, which combine research in Adaptive Hypermedia with traditional IR. The focus of DCM2 is on the metadata required by systems to provide this more intelligent behaviour. DCM2 includes work on generating, managing and linking structured knowledge in the form of ontologies. The main focus of this work is in addressing the shortcomings in current work on creating and sharing metadata between different intelligent systems. Finally, DCM3 focuses directly on the improvement of multi-modal Adaptive Hypermedia. This work includes applying AH techniques in tandem with multi-modal approaches, allowing, for example, for speech synthesis to be used where it improves the behaviour of the system.

Industry Engagement

CNGL DCM has engaged strongly with the industrial partners. There have been specific collaborations with Symantec and Microsoft under the framework of the Personalised Multilingual Customer Care demonstrator scenarios, with a view to incorporating real-world use cases and customer care content into the vision of next generation digital content management. Further collaboration has been undertaken with IBM, with a specific view towards the IBM LanguageWare technologies, including the LanguageWare workbench product and IBM Galaxy analysis tools. DCM has also continued to present to potential new CNGL industry partners, working with the centre management to attract further industry involvement. There are extensive plans to expand and deepen the industrial collaborations in DCM, including extending new scenarios for demonstration based on use cases driven from industrial needs, as well as locating new potential industrial partners who have specific digital content needs that can only be addressed by the CNGL DCM research team.