Graduate School and Research Center in Digital Sciences

Enriching media fragments with named entities for video classification

Li, Yunjia; Rizzo, Giuseppe; Redondo Garcia, José Luis; Troncy, Raphaël

WWW 2013, 1st Worldwide Web Workshop on Linked Media (LiME'13), May 13, 2013, Rio de Janeiro, Brazil

With the steady increase of videos published on media sharing platforms such as Dailymotion and YouTube, more and more efforts are spent to automatically annotate and or- ganize these videos. In this paper, we propose a framework for classifying video items using both textual features such as named entities extracted from subtitles, and temporal features such as the duration of the media fragments where particular entities are spotted. We implement four automatic machine learning algorithms for multiclass classification problems, namely Logistic Regression (LG), K-Nearest Neighbour (KNN), Naive Bayes (NB) and Support Vector Machine (SVM). We study the temporal distribution patterns of named entities extracted from 805 Dailymotion videos. The results show that the best performance using the entity distribution is obtained with KNN (overall accuracy of 46.58%) while the best performance using the temporal distribution of named entities for each type is obtained with SVM (overall accuracy of 43.60%). We conclude that this approach is promising for automatically classifying online videos.

Document Doi Bibtex

Title:Enriching media fragments with named entities for video classification
Keywords:Media Fragment, Video Classification, Media Annotation, Named Entity Extraction, Concept Extraction, NERD
Type:Conference
Language:English
City:Rio de Janeiro
Country:BRAZIL
Date:
Department:Data Science
Eurecom ref:3967
Copyright: © ACM, 2013. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in WWW 2013, 1st Worldwide Web Workshop on Linked Media (LiME'13), May 13, 2013, Rio de Janeiro, Brazil http://dx.doi.org/10.1145/2487788.2487970
Bibtex: @inproceedings{EURECOM+3967, doi = {http://dx.doi.org/10.1145/2487788.2487970}, year = {2013}, title = {{E}nriching media fragments with named entities for video classification}, author = {{L}i, {Y}unjia and {R}izzo, {G}iuseppe and {R}edondo {G}arcia, {J}os{\'e} {L}uis and {T}roncy, {R}apha{\"e}l}, booktitle = {{WWW} 2013, 1st {W}orldwide {W}eb {W}orkshop on {L}inked {M}edia ({L}i{ME}'13), {M}ay 13, 2013, {R}io de {J}aneiro, {B}razil}, address = {{R}io de {J}aneiro, {BRAZIL}}, month = {05}, url = {http://www.eurecom.fr/publication/3967} }
See also: