Social bookmarking tools are very popular nowadays. In most tools, users tag the bookmarks to describe them. Therefore, it is often hard for users to discover implicit structures between tags, users and bookmarks. We think that this is essential for both end users to discover new bookmarks that could be of interest to them, and for researchers who want to study how people use social information retrieval tools. In this work, a cluster map visualization application is customized to enable users to explore social bookmarks in the del.icio.us system [5]. The design of the application aims to automatically identify tag and community structures, and visualizes these structures in order to increase the users' awareness of them.
The ability to store or bookmark web page addresses has been one of the most important features of browsers since the beginning of the Web. Social bookmarking tools became possible when the process migrated from keeping the bookmarks on the client, to keeping them online on the Web, described by tags or terms. The purpose of social bookmarking tools is to tag the content of other users, mainly for the benefit of the tagger, although the bookmarks and tags are generally public, and users can establish networking opportunities [1]. The personal approach of tagging is an unstructured bottom-up approach of classifying content, in contrast to a top-down structured approach based on taxonomies, thesauri or ontologies. The semantic structures that result from a tagging approach are often referred to as a folksonomy.
The problem with tags is that they generally produce a flat namespace, rather than the hierarchical structures that taxonomies or other formal classification systems provide [1]. However, there can be rich implicit structures between tags, bookmarks and users. We designed a system that attempts to visualize these structures, so that end users can explore the social bookmarks in a playful, efficient and flexible way.
There are a number of social bookmarking tools, like CiteULike [2], Furl [3],
BlogMarks [4], etc. Del.icio.us [
We start this paper with a detailed overview of the design of our visualization in section 2. Section 3 describes the underlying technology. A short overview of related work is given in section 4 and we conclude the paper in section 5.
We designed a visualization application for the social bookmarks in del.icio.us. First we discuss the design requirements in section 2.1. In section 2.2, we look at the data we use for the application. The customized cluster map visualization is discussed in section 2.3 and we conclude with a typical exploration use case in section 2.4.
Our application for browsing social bookmarks was designed with a number of requirements in mind. First of all, we wanted to use and evaluate a novel paradigm for browsing social bookmarks, compared to the hypertext paradigm that is used on the del.icio.us website. Secondly, we wanted the system to be able to automatically identify tag and community structures. Such a structure is formed when two or more tags or users respectively describe or share common bookmarks. After the identification of these structures, they must be visualized for the users, so that they can become aware of them. We think that this is essential for both end users to discover new bookmarks that could be of interest to them, and for researchers who want to study how people use social information retrieval tools. On top of this, users should be able to explore the social bookmarks in a playful manner in a fun and engaging space. 2.2 Data
In order to visualize the content of del.icio.us, we have to collect the data first. We start from one or more del.icio.us user names and collect their bookmarks and the tags that describe them. Del.icio.us enables users to add other users to their network, so that they can access the bookmarks of these users. In the second step, we collect the tags and bookmarks for all the users that are in the network of the users identified in step 1. In a third step, the same data is collected for all the fans of the chosen user(s) in step 1. User A ‘is a fan of’ user B if user A has added user B to his network. This relation is not necessarily bi-directional.
During the exploration of this initial set of data, end users can ask the system to expand the data with the tags, bookmarks and network of users of their choice.
Tree Structure. This structure presents an overview of the tags and the users that are
currently loaded in the system. Fig. 2a shows that each tag is presented as a node in
the tree and the number of bookmarks, that are described with the tag, is indicated. In
the design of this tool, we choose to follow the philosophy of “start with what you
know, then grow” [
Bookmarks can be clustered by the users that have them in their collection, or by the tags that describe them. This can be seen in Fig. 4, where the tag ‘email’ is shown. Users ‘lisamac’ and ‘jgarber’ have one bookmark in common which is tagged by ‘email’. The user ‘jgarber’ has 2 extra bookmarks tagged with ‘email’ that are not in the collection of the other user. In Fig. 4., two colors are used to represent bookmarks – yellow and blue. The blue color means that the bookmark belongs to a selected set of bookmarks. Such a set can be created by clicking on a user or a tag in the tree pane, or by performing a keyword query in the search box. In Fig. 4, the blue bookmarks are also tagged with the tag ‘webdesign’ which is used to create a selection by clicking on this tag in the tree pane (Fig. 2a). The blue bookmark between ‘lisamac’ and ‘email’ is therefore in the collection of the users ‘lisamac’ and ‘jgarber’ and is described by the tags ‘email’ and ‘webdesign’. Result list. This list, shown in Fig. 2b shows detailed metadata about the bookmarks that are either selected in the cluster map, or bookmarks that match the search terms when a query was performed. The metadata cover the title, the location, user(s) that added the bookmark and tag(s) that describe the bookmark. A user can interact with the detailed information by clicking on e.g. a tag that describes a bookmark. If this tag is not already drawn on the cluster map, the visualization automatically updates itself and the tag classification is shown.
As an alternative to exploring social bookmarks on the del.icio.us webpage, we
offer a novel access paradigm that enables a user to explore social bookmarks in a
playful manner in a fun and engaging space. There are a number of ways that users
can start the exploration of the social bookmark space. First of all, a user can start
from an egocentric point-of-view, with a visualization of his bookmarks, much along
the lines of the philosophy of “start with what you know, then grow” [
A second way of exploring is browsing the tree structure to find interesting tags to visualize. Users can order the tag tree alphabetically or by the number of bookmarks they describe. By adding tags to the visualization, the corresponding bookmarks, possibly tagged by different users, can be explored in the cluster map.
Users can also interact with the visualization itself by clicking on nodes and
expanding these nodes and the visualization with new information. Upon clicking a
bookmark, the user can choose to show all associated tags. Depending on the choice,
the network and the fan data of all the users that saved this bookmark are loaded into
the system so that they become available for exploring. The reason that we do not add
the data of all users that saved a bookmark automatically from the start is the usage
throttling and abuse monitoring software at the del.icio.us website [
A last way of exploring del.icio.us social bookmarks with our tool is to enter one or more user names in a dialog box. All the data of those users is then loaded in the system and ready for exploring.
The tool for visually searching and analyzing del.icio.us social bookmarks is created
with the open and extensible information visualization framework that we created as
part of our research on the use of information visualization techniques for flexible and
efficient access to learning repositories [
For the visualization of the del.icio.us bookmarks, we plugged the Aduna Cluster
Map software [
With our framework, we can now also visualize the bookmarks as e.g. a tree-map where the bookmarks are classified per user (Fig. 5), in a flexible and efficient manner, or with any of the other techniques that our framework provides.
There a quite a few social bookmarking initiatives, like CiteULike [2], Connotea
[
All these initiatives are based on either visualizing the tag structures or the community structures where in our design both structures are taken into account for the visualization, so as to make more apparent implicit community and tag structures.
In this paper, we presented our design of a visualization application for social bookmarks. Our design tries to visualize implicit structures between tags, users and bookmarks. We customized a cluster map visualization technique for this purpose. The social bookmark tool del.icio.us was chosen as the source for the data to be visualized.
In the future, we will extend the system with data of other social bookmarking
tools. In this way, users can explore social bookmarks from multiple tools in our
application. Other possible data sources are folksonomies like Flickr [