Accurator Video

Socio-Economic Problem: Finding relevant multimedia content is notoriously difficult, and the difficulty increases with the size and heterogeneity of the content collection. Linked cultural media collections are heterogeneous by nature and rapidly increase in size, mainly through enormous amounts of user-generated content and metadata that are placed on the Internet on a daily basis. Without mechanisms for keeping any part of these collections easily accessible by any user at any time and any use context, the value of these collections for the community will drop, just like their value as an economic asset.

Scientific Problem: To solve the socio-economic problem stated above, the following key scientific/technological problems need to be addressed:

Combining machine-driven multimedia content labeling with the social content interaction concepts from Web 2.0 has been recognized in its potential to not only help improve the quality of labels assigned to content, but also to help enrich the interaction between the users and collections, e.g. in terms of personalization of content access. The social indexing concepts to be exploited include implicit and explicit HCI, individual and collaborative tagging, tag propagation and recommendation, as well as the relations between the users, content and tags. However, to date, this potential has not been tackled in any significant depth. The main research question is how different labelling and content interaction resources can be optimized and integrated together in a synergetic fashion to maximize content access reliability^2,3,4.
Up till now, semantic search techniques in a highly interlinked semantic search graph[1], based on RDF/OWL representations of enriched vocabularies, metadata, and hybrid search methods[2], combining ontology-based search and keyword-based matching to cope with the lack of semantic coverage of document content, have proved to be feasible. In this project we intend to further extend those approaches and develop a scalable and generalisable set search strategies by exploring further the benefits of “pattern-based” search[3], where the patterns are semantic-link structures in the graph. As our aim is also to provide techniques for clustering and ranking the search results in a for users intuitive fashion, we will step on personalization research, where content-based recommendation strategies[4] currently involve semantics and reasoning to discover additional knowledge about the user’s preferences and achieving more accurate personalization processes. The need for such a user-centered approach in the cultural domain has been justified[5].

The following key questions will be addressed to secure a successful realization of the project mission aiming to resolve the problems described above:

How to model and exploit the synergies between traditional multimedia labeling methods and semantically-relevant information inferred from users’ actions and behavior in social communities?
How to naturally and intuitively infer personal properties from individual user behavior and social interactions and to enrich them for efficient personalized search?
What are the critical Web interface design and user interaction issues stimulating active social engagement in sharing multimedia content in linked cultural media collections?
How to use the above to ensure uptake of the SEM paradigm by linked culture media collections?
How to develop optimal trust and access policies in case of linked protected curated content collections and free public Web data?

[1] J. Wielemaker, M. Hildebrand, J. van Ossenbruggen, G. Schreiber: Thesaurus-Based Search in Large Heterogeneous Collections, Int. Semantic Web Conference 2008

[2] R. Bhagdev, S. Chapman, F. Ciravegna, V. Lanfranchi, D. Petrelli: Hybrid Search: Effectively Combining Keywords and Semantic Searches, ESWC 2008

[3] L. Hollink, G. Schreiber, B. Wielinga: Patterns of Semantic Relations to Improve Image Content Search. Journal of Web Semantics, 2008

[4] Y. Blanco-Fernández, J.J. Pazos-Arias, A. Gil-Solla, M. Ramos-Cabrer, M. López-Nores: Semantic Reasoning: A Path to New Possibilities of Personalization. ESWC 2008

[5] S. Chan: Tagging and Searching – Serendipity and museum collection databases. In J. Trant and D. Bearman (eds). Museums and the Web 2007: Proceedings. Toronto: Archives & Museum Informatics