Paper about automatic labeling in IJDL

mompeltOur paper  “Evaluating Unsupervised Thesaurus-based Labeling of Audiovisual Content in an Archive Production Environment” was accepted for publication in the International Journal on Digital Libraries (IJDL). This paper, co-authored with Roeland Ordelman and Josefien Schuurman reports on a series of information extraction experiments carried out at the Netherlands Institute for Sound and Vision (NISV). Specifically, in the paper we report on a two-stage evaluation of unsupervised labeling of audiovisual content using subtitles. We look at how such an approach can provide acceptable results given requirements with respect to archival quality, authority and service levels to external users.

tess_alg

For this, we developed a text extraction pipeline (TESS), pictured here which extracts key terms and matches them to the NISV thesaurus, the GTAA. This journal paper is an extended version of the paper previously accepted at the TPDL conference and here provide an analysis of the term extraction after being taken into production, where we focus on performance variation with respect to term types and television programs. Having implemented the procedure in our production work-flow allows us to gradually develop the system further and to also assess the effect of the transformation from manual to automatic annotation from an end-user perspective.

The paper will appear on the Journal site shortly. A final draft version of the paper can be found here: deboer_ijdl2016evaluating_draft [PDF].

 

 

Advertisements

Dutch Ships and Sailors in 1st issue of the DHCommons journal

DHCommons journal logoA while ago, we submitted a project description of our Digital History project Dutch Ships and Sailors to the DHCommons journal and this week the first issue of the journal was published containing our paper “The Dutch Ships and Sailors project“.

This is a nice companion piece to the more technical description of the dataset which was published in the proceedings of ISWC 2014. The new version highlights more the general setup of the project and the considerations and innovations of the project from a historical point of view.

New datacloud

New datacloud

Since submission of this ‘mid-term project description’, the DSS data cloud has been expanding, and the ‘development’ version of the triple store now hosts six datasets thanks to the work of Jeroen Entjes (see the datacloud figure).