ABSTRACT
This paper presents a framework for categorizing channels of videos in a thematic taxonomy with high precision and coverage. The proposed approach consists of three main steps.First, videos are annotated by semantic entities describing their central topics. Second, semantic entities are mapped to categories using a combination of classifiers.Last, the categorization of channels is obtained by combining the results of both previous steps.
This framework has been deployed on the whole corpus of YouTube, in 8 languages, and used to build several user facing products. Beyond the description of the framework, this paper gives insight into practical aspects and experience: rationale from product requirements to the choice of the solution, spam filtering, human-based evaluations of the quality of the results, and measured metrics on the live site.
- Kurt D. Bollacker, Colin Evans, Praveen Paritosh, Tim Sturge, and Jamie Taylor. Freebase: a collaboratively created graph database for structuring human knowledge. In Proceedings of the SIGMOD Conference, 2008. Google ScholarDigital Library
- Razvan C. Bunescu, Marius Pasca, and Marius Pasca. Using encyclopedic knowledge for named entity disambiguation. In Proceedings of the EACL conference, 2006.Google Scholar
- Andras Csomai, Rada Mihalcea, and Rada Mihalcea. Linking documents to encyclopedic knowledge. 2008.Google ScholarDigital Library
- Silviu Cucerzan. Large-scale named entity disambiguation based on wikipedia data. In Proceedings of the EMNLP-CoNLL conference, 2007.Google Scholar
- Ofer Dekel, Joseph Keshet, and Yoram Singer. Large margin hierarchical classification. In Proceedings of the twenty-first international conference on Machine learning, ICML '04, New York, NY, USA, 2004. ACM. Google ScholarDigital Library
- Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Fei-Fei Li. Imagenet: A large-scale hierarchical image database. In Proceedings of the CVPR conference, 2009.Google ScholarCross Ref
- Stephen Dill, Nadav Eiron, David Gibson, Daniel Gruhl, Ramanathan V. Guha, Anant Jhingran, Tapas Kanungo, Sridhar Rajagopalan, Andrew Tomkins, John A. Tomlin, Jason Y. Zien, and Jason Y. Zien. Semtag and seeker: bootstrapping the semantic web via automated semantic annotation. In Proceedings of the WWW conference, 2003. Google ScholarDigital Library
- Olivier Duchenne, Ivan Laptev, Josef Sivic, Francis Bach, Jean Ponce, and Jean Ponce. Automatic annotation of human actions in video. In Proceedings of the ICCV conference, 2009.Google ScholarCross Ref
- D.A. Forsyth and J. Ponce. Computer Vision: A Modern Approach. (Second edition). Pearson Education Inc., 2011.Google Scholar
- Abhinav Gupta, Praveen Srinivasan, Jianbo Shi, Larry S. Davis, and Larry S. Davis. Understanding videos, constructing plots learning a visually grounded storyline model from annotated videos. In Proceedings of the CVPR conference, 2009.Google ScholarCross Ref
- Xianpei Han, Le Sun, Jun Zhao, and Jun Zhao. Collective entity linking in web text: a graph-based method. In Proceedings of the SIGIR conference, 2011. Google ScholarDigital Library
- Johannes Hoffart, Mohamed Amir Yosef, Ilaria Bordino, Hagen Furstenau, Manfred Pinkal, Marc Spaniol, Bilyana Taneva, Stefan Thater, Gerhard Weikum, and Gerhard Weikum. Robust disambiguation of named entities in text. In Proceedings of the EMNLP conference, 2011. Google ScholarDigital Library
- Sayali Kulkarni, Amit Singh, Ganesh Ramakrishnan, Soumen Chakrabarti, and Soumen Chakrabarti. Collective annotation of Wikipedia entities in web text. In Proceedings of the KDD conference, 2009. Google ScholarDigital Library
- Tie-Yan Liu, Yiming Yang, Hao Wan, Hua-Jun Zeng, Zheng Chen, and Wei-Ying Ma. Support vector machines classification with a very large-scale taxonomy. SIGKDD Explor. Newsl., 7(1), June 2005. Google ScholarDigital Library
- Marcin Marszalek, Ivan Laptev, Cordelia Schmid, and Cordelia Schmid. Actions in context. In Proceedings of the CVPR conference, 2009.Google ScholarCross Ref
- David N. Milne, Ian H. Witten, and Ian H. Witten. Learning to link with Wikipedia. In Proceedings of the CIKM conference, 2008. Google ScholarDigital Library
- Vicente Ordonez, Girish Kulkarni, Tamara L. Berg, and Tamara L. Berg. Im2text: Describing images using 1 million captioned photographs. In Proceedings of the NIPS conference, 2011.Google Scholar
- Yang Song, Ming Zhao, Jay Yagnik, and Xiaoyun Wu. Taxonomic classification for web-based videos. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2010.Google ScholarCross Ref
- Aixin Sun and Ee-Peng Lim. Hierarchical text classification and evaluation. In Proceedings of the 2001 IEEE International Conference on Data Mining, ICDM '01, Washington, DC, USA, 2001. IEEE Computer Society.Google ScholarDigital Library
- George Toderici, Hrishikesh Aradhye, Marius Pasca, Luciano Sbaiz, Jay Yagnik, and Jay Yagnik. Finding meaning on youtube: Tag recommendation and category discovery. In Proceedings of the CVPR conference, 2010.Google ScholarCross Ref
- Cornelis Joost "Keith" van Rijsbergen. Information Retrieval. Butterworth, London, Great Britain; Boston, Massachusetts, 1979.Google ScholarDigital Library
- Weilong Yang, George Toderici, and George Toderici. Discriminative tag learning on youtube videos with latent sub-tags. In Proceedings of the CVPR conference, 2011. Google ScholarDigital Library
- Zhicheng Zheng, Xiance Si, Fangtao Li, Edward Y. Chang, and Xiaoyan Zhu. Entity disambiguation with Freebase. In Proceedings of the IEEE/WIC/ACM International Conference on Web Intelligence, December 2012. Google ScholarDigital Library
Index Terms
- Classifying YouTube channels: a practical system
Recommendations
Editorial: Narrative-based taxonomy distillation for effective indexing of text collections
Taxonomies embody formalized knowledge and define aggregations between concepts/categories in a given domain, facilitating the organization of the data and making the contents easily accessible to the users. Since taxonomies have significant roles in ...
Balloon Synopsis: A Modern Node-Centric RDF Viewer and Browser for the Web
The Semantic Web: ESWC 2014 Satellite EventsAbstractNowadays, the RDF data model is a crucial part of the Semantic Web. Especially web developers favour RDF serialization formats like RDFa and JSON-LD. However, the visualization of large portions of RDF data in an appealing way is still a ...
Music on YouTube
We present the first study of YouTube's most popular content genre, music videos.Our analysis of popular music videos identified three main types and 12 subtypes.User-appropriated videos emerged as the most important new category of videos.Derivative ...
Comments