USER INTERFACE

The webLyzard information exploration and retrieval interface (=”dashboard”) builds upon a comprehensive repository of digital content, structured along geospatial, semantic and temporal dimensions. To visualize complex and often hidden relations within this repository, webLyzard integrates geographic maps, ontology graphs, tag clouds, information landscapes (visual representations of document clusters), and just-in-time information retrieval agents suggesting documents about similar topics and nearby locations. Geospatial technologies are crucial for this integration – they are not only a natural way to show the most relevant documents in their regional context, but also excellent platforms to deliver interactive visualizations of non-geographic data. Such visualizations help users to perceive and understand context when navigating large archives of Web content, and to analyze the perceptions and informational needs of various stakeholders.

Interface Structure

Drop-down elements in the main menu let users choose a time interval, the document source (news media, blogs, social media such as Twitter and Facebook, etc.), and the global sentiment filter setting (unfiltered, positive, negative). In conjunction with a full-text search and the resulting keyword graph, the global sentiment filter is an effective means to investigate a stakeholder group’s associations with a topic of interest. Below the main menu, users have access to the following interface elements:

• The trend chart in the upper left corner shows keywords and trends (frequency, sentiment) in the previous two months.
• The content view below shows the active document including keywords, publication date, place of publication (source geography), and primary location that is being referenced (target geography).
• Further below, just-in-time information retrieval agents list documents referring to similar topics and nearby locations.
• Visual components on the right side of the screen show various dimensions of the annotated documents. The information retrieval agents as well as the maps can be repositioned using drag-and-drop operations.

Date Selector

Users can adjust the time interval and access historic data by selecting “from” and “to” dates using the two calendar elements. This is a global setting that not only affects the trend chart, as outlined in the next section, but also limits search queries and dynamic visualizations to the chosen time interval. Dates can be selected by navigating one day or week into the past or future by using the provided back and forward buttons, or by selecting a specific date using the drop-down calendar. Selecting a new “to” date automatically updates the trend charts, the active document, as well as the semantic map in case the new date belongs to a different weekly snapshot.

Real-Time Map Synchronization

The maps on the right side facilitate access to the underlying knowledge base. Clicking on the ‘maximize’ button increases the size of the maps; clicking on the ‘popup’ button opens the map in a separate browser window (which allows using the system in multiple-screen configurations). Maps can be rearranged by dragging them to the desired position, and switched on and off using the buttons in the top bar.

The various windows are tightly coupled – user actions in one window trigger an immediate update of all other displays. As an alternative to entering query terms (see previous section) to find specific documents, users can click on any position in the maps (not only on the markers) to retrieve articles related to that particular location, topic or domain concept.

Hovering above a map previews the document closest to the current position of the mouse pointer, but does not activate it. When previewing documents, the other visualizations on the right side automatically adjust to show the immediate context of the previewed documents. Users who want to focus on a particular region can disable this default setting and “freeze” the currently displayed part of the map by pressing the pause button.