Web 2.0 dictionary

How might we benefit from the billions of tagged multimedia files (e.g. image, video, audio) available on the Internet? This paper presents a new concept called Web 2.0 Dictionary, a dynamic dictionary that takes advantage of, and is in fact built from, the huge database of tags available on the Web. The Web 2.0 Dictionary distinguishes itself from the traditional dictionary in six main ways: (1) it is fully automatic because it downloads tags from the Web and inserts this new information into the dictionary; (2) it is dynamic because each time a new shared image/video is uploaded, a "bag-of-tags" corresponding to the image/video will be downloaded, thus updating Web 2.0 Dictionary. The Web 2.0 Dictionary is literally updating every second, which is not true of the traditional dictionary; (3) it integrates all kinds of languages (e.g. English, Chinese), as long as the images/videos are tagged with words from such languages; (4) it is built by distilling a small amount of useful information from a massive and noisy tag database maintained by the entire Internet community, therefore the relatively small amount of noise present in the database will not affect it; (5) it truly reflects the most prevalent and relevant explanations in the world, unaffected by majoritarian views and political leanings. It is a real, free dictionary. Unlike Wikipedia" [5] which can be easily revised by even a single person, the Web 2.0 Dictionary is very stable because its contents are informed by a whole community of users that upload photo/videos; (6) it provides a correlation value between every two words ranging from 0 to 1. The correlation values stored in the dictionary have wide applications. We demonstrate the effectiveness of the Web 2.0 Dictionary for image/video understanding and retrieval, object categorization, tagging recommendation, etc, in this paper.

[1]  Marcel Worring,et al.  Content-Based Image Retrieval at the End of the Early Years , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  James Ze Wang,et al.  The Story Picturing Engine---a system for automatic text illustration , 2006, TOMCCAP.

[3]  Anton Nijholt,et al.  Prediction Strategies in a TV Recommender System – Method and Experiments , 2003, ICWI.

[4]  Yixin Chen,et al.  CLUE: cluster-based retrieval of images by unsupervised learning , 2005, IEEE Transactions on Image Processing.

[5]  Shih-Fu Chang,et al.  Image Retrieval: Current Techniques, Promising Directions, and Open Issues , 1999, J. Vis. Commun. Image Represent..

[6]  Tao Mei,et al.  Online video recommendation based on multimodal fusion and relevance feedback , 2007, CIVR '07.

[7]  Irena Koprinska,et al.  INTIMATE: a Web-based movie recommender using text categorization , 2003, Proceedings IEEE/WIC International Conference on Web Intelligence (WI 2003).

[8]  Gang Wang,et al.  Object image retrieval by exploiting online knowledge resources , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Antonio Criminisi,et al.  Harvesting Image Databases from the Web , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[10]  Pietro Perona,et al.  Learning object categories from Google's image search , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[11]  Andreas Stafylopatis,et al.  A hybrid movie recommender system based on neural networks , 2005, 5th International Conference on Intelligent Systems Design and Applications (ISDA'05).

[12]  David A. Forsyth,et al.  Animals on the Web , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[13]  James Ze Wang,et al.  Real-Time Computerized Annotation of Pictures , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Pietro Perona,et al.  A Visual Category Filter for Google Images , 2004, ECCV.