The project will be based on complementary and highly interconnected work coming from different disciplines. The five broad areas outside Computer Science which will contribute to the project are:

1) Philosophy of Language and Science, Cognitive Science

2) Library Science

3) Semiotics

4) Political Science

5) Science of Campaigns

The first two will provide the foundations for the study and use of context in the representation of knowledge. Library science will provide the foundations and experience needed to organize information in categories and to realize innovative mechanisms for indexing hierarchical categorisation schemes with meaningful concept sequences, i.e., Facets. Semiotics will allow for the definition of modal approaches to the discovery of knowledge diversity and of how web components and multimedia data can be used to express opinions and bias. The Science of Campaigns will transport the planned success in content analysis into the art of communicating the message in the web and media landscape, where images in politics and the economy fight for their preferred position, and connect the developed tool with every day communication practice.

Furthermore, the development of a system able to make computational sense of the theories developed in these interdisciplinary areas will require the integration of different sub-disciplines of Computer Science which are very rarely integrated. For instance, the implementation and use of context as the main means to represent diversity and the implementation and deployment of facets will require the integration of competences from Machine Learning, Information Retrieval (IR) and Knowledge Representation. For example, in traditional IR, a retrieval decision is generally made on the basis of the query and document collection, ignoring the user and the search context [Shen et al. 2005]¹. But information about the context can be gleaned from a number of areas of computing, including cognitive science accounts of the cognitive viewpoint, human-computer interaction, and natural language processing and knowledge acquisition, amongst others [Ingwersen and Jarvelin 2005]². Similarly, competences and technologies in pattern recognition and computational linguistics will need to be integrated in order to extract and manage opinions, bias and diversity based on an effective semiotic approach to text combined with images. It must also not be overlooked that context itself is not predefined or fixed, and has a dynamic of its own [Coutaz et al. 2005]³. Context must be discovered and re-negotiated constantly as part of a process of interaction with distributed, multiscale and reconfigurable resources.

1. X. Shen, B. Tan and C. Zhai. Context-sensitive information retrieval using implicit feedback. Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval, 43-50, 2005. [↩]
2. P. Ingwersen and K. Jarvelin. The Turn: Integration of Information Seeking and Retrieval in Context. Springer, 2005. [↩]
3. J. Coutaz, J.L. Crowley, S. Dobson and D. Garlan. Context is key. Communications of the ACM 48(3), 49-53, 2005. [↩]