Introduction email and use case proposition

Hi!

I am a postdoc at the VU University of Amsterdam [1]; after doing my PhD in
collaboration with the French National Institute for Audiovisual documents (the
French national AV archives), I have been working on a Dutch project co-hosted
at the Dutch Institute for Sound and Vision (the Dutch national AV archives),
aiming at generating annotation suggestions from texts for the documentalists of
this Institute. The annotation schema that we follow is the one defined at Sound
and Vision, in their current catalogue called iMMiX. More information about the
project can be found at [2]. 

My field of expertise is the Natural Language Processing, I have now close to 6
years of working experience with the annotation of audiovisual archives as an
applicative context for my research (always on the text/concept part, I’m not a
“signal person”, I’ve never worked on algorithms applied to the video stream
itself,only to textual resources related to video content), and for almost 3
years I am in a group deeply involved in the Semantic Web, so I have built some
knowledge on that side too.

The use cases I was thinking of are related to audiovisual archive as a Cultural
Heritage institution: 
-	a normalized content annotation will lead to improved
exchange/matching/merging possibilities, to show together different aspects of
the same person/place/event represented in different format and with different
information in different collections. The collections could be different
audiovisual archives, radio archives or musea.
-	A normalized representation of the semantic of the content of audiovisual
documents will enable a more systematic, efficient automatic linking with
external resources, as DBpedia [3] for example.
Of course, these are rough ideas, all comments (including criticism) are welcome!

Best regards,
Veronique

[1] http://www.few.vu.nl/~vmalaise
[2] http://www.nwo.nl/CATCH/CHOICE
[3] http://dbpedia.org/About

Received on Friday, 12 September 2008 17:20:44 UTC