Activities in the research and technology cluster will investigate new means for increasing the tacit productivity of individual knowledge workers, their communities and thus the overall productivity of enterprises. Learning, adapting, and exploiting contexts will play a crucial role in tailoring the knowledge workspace to the current needs of knowledge workers. These activities will cover both, research and development of prototypes that are integrated into the overall knowledge workspace application.

Research aspects in the four main fields of the project, which main dependencies are shown in the figure above, will be covered in Workpackages 1 – 4:

• The development of methods and tools for collaborative articulation and sharing of enterprise knowledge. ACTIVE systems will allow the articulation of expressive knowledge in a collaborative, flexible and intuitive way, e.g. by leveraging tools from the Web 2.0 landscape as in initial activities such as for example Semantic Media Wiki
• The development of techniques for spotting expedient refinements to initial (rather lightweight) knowledge structures and leveraging the knowledge base accordingly. Similarly, ACTIVE systems will detect and repair the likely inconsistencies in collaboratively maintained knowledge bases, based on improved metadata evolution methods. Among others, it will be investigated to which extent initial core ontologies will provide added-value in these combined top-down/bottom-up techniques
• The mediation between different formalisms and annotation schemas, both on the Semantic Web side, as well as on the Web 2.0 side

• The development of automatic and semiautomatic methods for identification the structure and dynamics of temporal processes through modern analytic methods. The idea is to bootstrap with or without human intervention analytic models which model complex behaviour within collaborative knowledge processes exploiting multi modal data including: structured data, text, social networks, visual information, and possibly sensor data – the goal is to be able to build models by combining different modalities
• The provision of human understandable explanations of inquiries about the modelled processes (e.g. by generating human readable reasoning traces or by visualization of complex data relationships) and efficient recommendations for improving the modelled processes (e.g. improving the structure of collaboration to make processes more efficient). A second learning task in WP2 is the learning of context information that is related to these knowledge processes
• The generation of user-specific contexts by exploiting tagging information or observing user behaviour, and the dynamic adaptation of these contexts to changing user needs

• The development of context-aware adaptation of the knowledge workspace for the current user
• The development of recommendation facilities that suggest to the knowledge worker which information is best to use for a specific task, e.g. based on community experience and filtering mechanisms
• The development of tools to predict future actions and behaviour within the modelled processes for efficient planning
• The development of monitoring and optimisation techniques for the knowledge process execution to improve the available context information and processes
• The identification of security or policy violations during knowledge process execution
• The development of adaptation methods that allow for user-centred filtering of relevant knowledge. ACTIVE users will use simple metaphors to design adaptation rules and policies. The use of easy-to-use mechanisms will – similarly to the case of initial knowledge articulation – ensure a low overhead in communicating the desired adaptation behaviour of the system

• The development of reliable methods to assess the costs of core Web2.0 and semantic technological solutions such as folksonomies, (lightweight) ontologies or tagging systems in order to demonstrate their tangible and measurable total benefits within enterprises as an argument to encourage their adoption. These methods will be used to support and optimize collaborative knowledge creation processes, which are covered by WP1, WP2 and WP3
• The study of incentives structures appropriate for ACTIVE technology. The results of this study will be available at an early stage of the project so that they can be taken into account at design time by the technology work packages WP1, 2, 3 and 5

The research-oriented workpackages above will be supplemented by workpackages 5-6 in developing industry-strength software modules, namely:

Knowledge Workspace (WP5)

• Will create a common software architecture to bind software components developed in WP1-WP4 and software developed and used by case studies in WP8-WP10 in a consistent way. WP5 will focus on providing the envisioned context-driven knowledge workspace, with adequate user interfaces and integration capabilities for desktop applications as well as enterprise-wide management of user profiles, contexts, processes etc. WP5 will define how to create ACTIVated tools by specifying both APIs and the ACTIVation process. WP5 will deliver this in form of a software development kit (ACTIVE SDK) that will be used in other WPs to develop ACTIVated software components.

• Will focus on extending various software tools like personal productivity tools and knowledge management tools used in enterprises to fit into the ACTIVE Knowledge Workspace by using the ACTIVE SDK. Such integration is particularly useful for knowledge workers who are used to interact with standard software such as the Microsoft Outlook PIM (Personal Information Management) system or the Mozilla Firefox Browser. Our analysis shows that Microsoft office tools from the 2007 edition provide very reach set of interfaces to develop various plugins/extensions. For example, Outlook 2007 SDK exposes entire outlook object model and provides hundreds of interface functions. This way it is possible to develop an Outlook plugin which will host the ACTIVE SDK and new ACTIVE–specific features for outlook. Internet explorer and Windows Desktop Search tools are also highly extendible through a rich set of APIs. For open source tools there is always a possibility to extend the original product source code, but for tools like Mozilla Thunderbird and Firefox there is also a number of APIs that can be used to develop plugins and extensions which can host the ACTIVE SDK and new ACTIVE–specific features.