Hierarchical Topic Models for Language-based Video Hyperlinking

Anca-Roxana Simon 1 Rémi Bois 1 Guillaume Gravier 1 Pascale Sébillot 1 Emmanuel Morin 2 Sien Moens 3
1 LinkMedia - Creating and exploiting explicit links between multimedia fragments
IRISA-D6 - MEDIA ET INTERACTIONS, Inria Rennes – Bretagne Atlantique
2 TALN
LINA - Laboratoire d'Informatique de Nantes Atlantique
3 Department of Computer Science - K.U.Leuven
Abstract : We investigate video hyperlinking based on speech transcripts , leveraging a hierarchical topical structure to address two essential aspects of hyperlinking, namely, serendipity control and link justification. We propose and compare different approaches exploiting a hierarchy of topic models as an intermediate representation to compare the transcripts of video segments. These hierarchical representations offer a basis to characterize the hyperlinks, thanks to the knowledge of the topics who contributed to the creation of the links, and to control serendipity by choosing to give more weights to either general or specific topics. Experiments are performed on BBC videos from the Search and Hyperlinking task at MediaEval. Link precisions similar to those of direct text comparison are achieved however exhibiting different targets along with a potential control of serendipity.
Keywords : hyperlinking topic modeling Multimedia NLP
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [10 references]  Display  Hide  Download
https://hal.archives-ouvertes.fr/hal-01186429
Contributor : Guillaume Gravier <>
Submitted on : Wednesday, September 2, 2015 - 1:18:41 AM
Last modification on : Thursday, February 21, 2019 - 10:31:46 AM
Document(s) archivé(s) le : Wednesday, April 26, 2017 - 10:17:16 AM

File

simon-slam01.pdf
Files produced by the author(s)

Identifiers

Collections

UNIV-RENNES1 | LINA | INSTITUT-TELECOM | UNIV-NANTES | INRIA | IRISA | IRISA-INSA-R | IRISA-D6 | LINA-TALN | UR1-UFR-ISTIC | CENTRALESUPELEC | INSA-RENNES | UNIV-RENNES

Citation

Anca-Roxana Simon, Rémi Bois, Guillaume Gravier, Pascale Sébillot, Emmanuel Morin, et al.. Hierarchical Topic Models for Language-based Video Hyperlinking. Workshop on Speech, Language and Audio in Multimedia, 2015, Brisbane, Australia. ⟨10.1145/2802558.2814642⟩. ⟨hal-01186429⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

2057

Files downloads

244

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD