Visualizing tags over time

Abstract

We consider the problem of visualizing the evolution of tags within the Flickr (flickr.com) online image sharing community. Any user of the Flickr service may append a tag to any photo in the system. Over the past year, users have on average added over a million tags each week. Understanding the evolution of these tags over time is therefore a challenging task. We present a new approach based on a characterization of the most interesting tags associated with a sliding interval of time. An animation provided via Flash in a web browser allows the user to observe and interact with the interesting tags as they evolve over time.New algorithms and data structures are required to support the efficient generation of this visualization. We combine a novel solution to an interval covering problem with extensions to previous work on score aggregation in order to create an efficient backend system capable of producing visualizations at arbitrary scales on this large dataset in real time.

Exploring Characteristics of Social Classification

“…Partington's (2004) analysis suggested categories such as medium, subject, genre, name and location. Dubrinko's (2006) The set of categories (and their brief definitions) developed for this study is shown in Figure 2. It was believed that these categories best reflect the tags employed on the flickr.com site for images.…”

User-Defined Classification on the Online Photo Sharing Site Flickr…or, How I Learned to Stop Worrying and Love the Million Typing Monkeys

“…For example, in February 2005, 62% of the images stored on Flickr, 2.74 million images, had at least one tag (Butterfield, 2005). And over the last year, Flickr users have added, on average, over one million tags per week to the dataset (Dubinko, 2006). In addition to assisting in organization and indexing, Flickr's tagging tool allows users to build communities and unique narrative experiences around their images, define their own individual organizational structure on top of the standard base content, and to explore personal image clusters which are query defined and highly malleable.…”

Disasters seen through Flickr cameras

“…However, tags that significantly describe the singularities of the reported event are those occurring abundantly in t and rarely outside of t. In order to find these significant tags we take advantage of the notion of interestingness introduced by Dubinko et al [2].…”

“…Similarly to the tf-idf weighting scheme from information retrieval [7], the interestingness as presented by Dubinko et al [2] aims at exhibiting the most significant object that occurs in a subset even though it is not necessarily the most frequent one. To this end the measure takes into account the number of occurrences of a tag during a particular time interval and during the total studied period of time.…”