This article explores the task of linking written and audiovisual News, based on the use of semantic textual similarity metrics. It presents a comprehensive study of different linking approaches with various configurations of inter-media or intra-media association. The influence of document length and request length is also explored. It is shown that textual similarity metrics that have proved to perform very well in the context of community question answering can provide efficient News linking metrics, whatever the media association configuration.
@InProceedings{CHARLET18.5, author = {Delphine Charlet and Géraldine Damnati}, title = {Linking written News and TV Broadcast News topic segments with semantic textual similarity}, booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {may}, date = {7-12}, location = {Miyazaki, Japan}, editor = {Octavian Popescu and Carlo Strapparava}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, isbn = {979-10-95546-11-5}, language = {english} }
Summary of the paper