The recent advent and wide adoption of Social Bookmarking Systems (SBS) has disrupted the traditional model of online content publishing and consumption. Until recently, the majority of content consumed by people was published as a result of a centralized selection process. Nowadays, the large-scale adoption of the Web 2.0 paradigm has diffused the content selection process to the masses. Modern SBS-based applications permit their users to submit their preferred content, comment on and rate the content of other users and establish social relations with each other. As a result, the evolution of popularity of socially bookmarked content constitutes nowadays an overly complex phenomenon calling for a multiaspect analysis approach. This chapter attempts to provide a unified treatment of the phenomenon by studying four aspects of popularity of socially bookmarked content: (a) the distributional properties of content consumption, (b) its evolution in time, (c) the correlation between the semantics of online content and its popularity, and (d) the impact of online social networks on the content consumption behavior of individuals. To this end, a case study is presented where the proposed analysis framework is applied to a large dataset collected from Digg, a popular social bookmarking and rating application.
TopIntroduction
The emergence of Web 2.0 technologies and the widespread use of applications integrating such technologies have transformed the way people experience and act in online settings. In the first days of the Web, people were excited to browse through and consume online content (mostly static web pages) that was prepared and published by website owners or administrators. Nowadays, digital content consumption – e.g. online article reading, picture viewing and video watching – still appears to be one of the main activities for most internet users1. However, the advent of the Web 2.0 application paradigm has transformed the established “browsing-based” online content consumption behavior of users. This change was possible by means of offering users a host of rich interactivity features, such as content sharing, rating as well as online community building. Thus, users of today’s Web 2.0 applications are empowered to share, organize, rate and retrieve online content. In addition, users are exposed to the content-related activities of other users and can even form online relations to each other. Consequently, online content consumption within a modern Web 2.0 application constitutes an overly complex phenomenon with interesting dynamics which have not been thoroughly investigated yet.
Social Bookmarking Systems (SBS) hold a prominent place among Web 2.0 applications with respect to content consumption since they provide a platform where users are provided with two significant features:
- •
Submitting and sharing bookmarks (links) to online resources, e.g. articles, photos or videos, which they consider interesting.
- •
Indicating their preference or disapproval to bookmarks submitted by other users, by voting for or against the interesting-ness/appeal of online resources and by commenting on them.
In addition to these two features which are fundamental for an SBS, there are two other optional groups of features, namely Taxonomic and Social-Community features. Taxonomic features pertain to the possibilities offered to users for assigning bookmarks to a predefined topic-scheme or for tagging them with freely chosen keywords. Social-Community features enable the users to create “friendship” relations with each other (which can be unilateral or mutual) or to create groups of topical interests. A number of social bookmarking applications have been recently launched. Systems such as delicious are meant to be used as general bookmark organization and sharing applications, while there are also social bookmarking applications focused on online news such as Digg and newsvine, and even niche bookmarking services such as CiteULike, used only for bookmarking citations to research articles. Table 1 lists some of the most popular SBS along with their features.
There has been a recent surge in the usage of social bookmarking applications, many of which currently attract several million of unique users per month. An illustrative overview of usage statistics pertaining to the most popular SBS is provided in Table 2, where the top 10 social bookmarking applications are ranked based on the number of unique monthly visitors they attract. Several additional web popularity metrics are provided in the same table, namely the number of inbound links, the Google Page Rank and the Alexa Rank. According to the table, Digg, StumbleUpon, propeller and reddit are the most popular social bookmarking applications in terms of the number of unique visitors per month. Although delicious is the oldest bookmarking application from this list and features the largest number of bookmarks, it appears to attract a smaller number of visitors than its main competitors.
The widespread and intense use of such applications is responsible for the ceaseless creation of massive data sets where the content consumption patterns of users are imprinted. The description and analysis of the intricate phenomena related to these patterns could provide answers to a series of interesting questions: