Extracting Time and Location Concepts Related to Tags

Folksonomy is a method of classifying content, and it is widely used in some web services. It allows users to choose tags (keywords or terms assigned to specific content) freely and to search content by referring to the tags. Compared to existing classification methods, folksonomy reflects the users’ intention more directly because of its unlimited vocabulary and multiple tags for one content item. Moreover, it has a useful characteristic where tags represent the description of the content. Although tags are intended to be a rich semantic description of web content, machines cannot understand what the tags mean because they are just keywords. We describe a method to extract the concept related to the tag in a machine-understandable way by focusing on the features of content annotated with each tag. In particular, we target the problem of extracting the temporal and spatial concepts of the tags on Flickr, a popular photo sharing service by looking at the date and location distributions of photos for each tag. We evaluated the concept extracting method on a snapshot of actual Flickr data and show that it can identify a tags’ concept in a manner similar to the way a person can.

[1]  Yiu-Fai Wong,et al.  Clustering Data by Melting , 1993, Neural Computation.

[2]  Ravi Kumar,et al.  Visualizing tags over time , 2006, WWW '06.

[3]  Mor Naaman,et al.  World explorer: visualizing aggregate data from unstructured text in geo-referenced collections , 2007, JCDL '07.

[4]  Roelof van Zwol,et al.  Flickr tag recommendation based on collective knowledge , 2008, WWW.

[5]  Thomas Gruber,et al.  Ontology of Folksonomy: A Mash-Up of Apples and Oranges , 2007, Int. J. Semantic Web Inf. Syst..

[6]  Andrew P. Witkin,et al.  Scale-Space Filtering , 1983, IJCAI.

[7]  Oscar Firschein,et al.  Readings in computer vision: issues, problems, principles, and paradigms , 1987 .

[8]  Yong Yu,et al.  An Unsupervised Model for Exploring Hierarchical Semantics from Social Annotations , 2007, ISWC/ASWC.

[9]  Mor Naaman,et al.  HT06, tagging paper, taxonomy, Flickr, academic article, to read , 2006, HYPERTEXT '06.

[10]  P. Schmitz,et al.  Inducing Ontology from Flickr Tags , 2006 .

[11]  Grigory Begelman,et al.  Automated Tag Clustering: Improving search and exploration in the tag space , 2006 .

[12]  Peter J. Nürnberg,et al.  Proceedings of the seventeenth conference on Hypertext and hypermedia , 2006 .

[13]  Enrico Motta,et al.  Bridging the gap between folksonomies and the semantic web: an experience report , 2007 .

[14]  Bernardo A. Huberman,et al.  The Structure of Collaborative Tagging Systems , 2005, ArXiv.

[15]  Jianchang Mao,et al.  Towards the Semantic Web: Collaborative Tag Suggestions , 2006 .

[16]  Mor Naaman,et al.  Why we tag: motivations for annotation in mobile and online media , 2007, CHI.

[17]  Yee Leung,et al.  Clustering by Scale-Space Filtering , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  Mor Naaman,et al.  Towards automatic extraction of event and place semantics from flickr tags , 2007, SIGIR.

[19]  Dimitrios Gunopulos,et al.  Identifying similarities, periodicities and bursts for online search queries , 2004, SIGMOD '04.

[20]  Joydeep Ghosh,et al.  Scale-based clustering using the radial basis function network , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).