A Synchronized Tag Cloud and Timeline Visualization Joris Klerkx and Erik Duval Abstract—In this paper, we present a synchronized Tag Cloud and Timeline Visualization, that is created to enable users and researchers to get an insight in the evolving patterns and focus of their interests while searching the network of repositories within the Global Learning Objects Brokered Exchange (GLOBE) consortium. The visualization is an interactive SVG drawing that is synchronized with timeline visualization by using JavaScript. Index Terms—tag clouds, timeline, visualization, attention metadata, web search history 1 INTRODUCTION We created a web application (GlobeMash [1]) that enables users to search the network of repositories within the Global Learning Objects Brokered Exchange (GLOBE) consortium [1]. Attention Metadata, a way to represent data about the activities of users in a certain environment [3], is tracked within GlobeMash to collect the search history of end users. In this paper, we describe a browser- based application to visualize the evolution of search keywords within GlobeMash. The goal of this application is to enable end users and researchers to get an insight in the evolving patterns and focus of their interests. We start this paper with a description of why we want to visualize the web search history of users in GlobeMash. In section 3, we discuss the visualization in detail. Section 4 gives an overview of related work and we conclude this paper is section 5. 2 WEB SEARCH HISTORY In order to create a feedback loop that enables learning from the way people actually use the search interface within GLOBE, it is essential to track the attention and behavior of users within GlobeMash. The tracking information is needed to answer questions like • Which keywords are used how many times and could therefore be interpreted as representative keywords for a time period? • Which keywords are used that have many or no results? • Which keywords have results in multiple repositories? • Which keywords or topics have the attention of a particular user over time? • What are the behavior and the focus of the community within the GLOBE? • How many results does a user look at in detail after performing a query? • When did a user perform a query and which keywords did he or she use? • What is the evolution of search keywords over time? • Etc. These questions are relevant for the end users and researchers that are interested to get an insight in the evolving patterns and focus of the users’ interest, or to use the users’ search history for e.g. recommendation purposes or to establish social networking opportunities. Every search event in GlobeMash is therefore logged in Contextualized Attention Metadata (CAM) [3]. A visualization of this data is created to enable users to get an insight in the questions above in a flexible, efficient and browser-based way. 3 VISUALIZATION Our visualization is made up of two synchronized parts – an extended tag cloud and a timeline that indicates the different events that are tracked by the Attention Metadata and the current interval of time that is visualized in the tag cloud. These parts can be seen in Figure 1. In the next paragraphs we describe these parts in detail. Fig. 1. A Synchronized Tag Cloud and Timeline. 3.1 Tag Cloud Tag Clouds are visual presentations of a set of words, typically a set of tags, in which attributes of the text such as text, weight or color can be used to represent features (e.g. frequency) of the associated terms [4]. When a user clicks on a tag, the user obtains an ordered list of tag-described resources, as well a list of related tags [5]. Usually, the tags are displayed in alphabetical order. In our visualization, every search keyword that is used in GLOBE is represented as a tag in the Tag Cloud. However, tags are not ordered alphabetically, but in a coordinate system where the X-axis represents the average number of search results and the Y-axis the number of repositories where results were found, both in relation to the performed keyword query. The scale of the x-axis and the y-axis are automatically computed from the data that has to be drawn in the visualization. On top of that, the visualization offers the functionality to draw circles around a tag or keyword. The radius of the circle represents the frequency of the keyword used in GlobeMash over time. The color-coding of these circles can be random or it can be used to • Joris Klerkx and Erik Duval are with the Dept. ComputerScience, Katholieke Universiteit Leuven, Celestijnenlaan 200A, B-3001 Leuven, Belgium • Email: {Joris.klerkx, Erik.duval}@cs.kuleuven.be .