Cluster Map
Search results are visualized as overlapped document clusters in the "Cluster Map" panel. Each cluster represents a single term in current search query and consists of a number of documents matching this term. When adding new term to a search query (either via keyword search, suggested term selection, or selection of a facet value), the cluster map is regenerated to display how the term is related to others via documents they have in common.

The figure shows the Cluster Map after the user has evaluated three queries: a keyword search for the term "search", search for the phrase "knowledge management" and a search for all PDF documents (using the 'Type' facet). The documents resulting from these searches are shown as spheres, with colors corresponding to the source type: yellow spheres represent files; green spheres are web pages, etc. Each of the four document sets is symbolized by a rounded rectangle, labeled with the facet value and an icon indicating the facet. The colored edges rendered in the background connect documents to labels, indicating that these documents belong to that result set. A document contained in multiple result sets is displayed only once but connected to the labels of the sets that contain it. Subsets of documents with the same set memberships are placed together, thus clustering the total set of documents. When a cluster contains more than 250 documents, it is displayed as a cylinder labeled with the number of documents, to prevent very large clusters from disturbing the usability of the visualization. When the graph becomes too big to display in the available space in its natural size, the visualization is automatically scaled so that the graph is always visible in its entirety.
Details about individual results are shown in a tooltip that pops up when the user hovers the mouse pointer over a document in a cluster:

Double-clicking a document item will open the document with specific external application (see "Opening the documents" below).
The terms contained in a current query are listed in the box at top right corner of the Cluster Map panel. To exclude a specific term from the cluster map, click on Remove (X) icon for this term in the list.
To clear the cluster map (remove all terms from it) and start new search, press "Clear" button in the terms list.
If the cluster map regeneration takes too long, you can stop the process by pressing "Stop" button.
Detailed results view
In order to inspect the contents of the visualization, the user can select a cluster or result set by clicking on it. Its contents will be displayed in the "Details" panel below the map. This panel contains a list of the documents that can be presented in two modes:
- Table view displays the results as table where each row represents a single document and the colums represent the document attributes such as title, date, location etc. The set of the attributes to display can be customized with "Toggle visible table columns" menu of the Details panel. Click on a table column header to sort the table by specific document property.
- List view displays the results in a form, similar to the web search engines. For each document, the title and other important metadata is displayed. For user's convenience, the list is splitted to separate pages and allows to navigate between the result pages.
Opening a document
Documents can be opened in their native application by double-clicking on their spheres in the graph, on their table entries or on their titles (in the list view mode).
Please note: Depending on your platform and system configuration, you may be asked for a path to the application executable in a dialog box.
Using significant terms
When the user selects documents in the Cluster Map, statistical measures are used to determine relevant terms for this set. These terms are listed as suggestions in the Navigation panel (see Search) and can be used to narrow down the search. When no selection is made, the union of all document sets is used in the determination of the suggestions.
Significant terms for specific documents are displayed either in "Significant Terms" column of the Table view, or at a bottom line of a corresponding List view item. To add a term to the query, click on the "Significant terms" column for a specific document and select the term in a drop-down list. In the case of the List View mode, click on a term of interest.
Visualization options
Cluster map options
The cluster map options and settings are represented with the buttons in the Cluster Map panel toolbar.
- Display individual results / Display clusters as single visual entities These two buttons switch between two modes of cluster visualisation. In the first mode, each individual document is shown as an individual visible particle within a cluster. Hovering the mouse cursor above these particles, a user can get information about a document displayed as a tooltip. In the second mode, individual documents are not displayed within the clusters.
- Change color scheme The button opens a menu to choose a scheme to colorize the clusters in the map.
- Hide less connected clusters If the option is on, the clusters with less number of connections will be sorted out of the map.
- Scale the graph to fit the available screen space If the option is on, the cluster map size will be automatically adjusted for the screen space of the panel (no scrollbars are appeared). This option is duplicated with the "View → Cluster Map → Scale to fit window" menu item.
Export Cluster Map
You can export current cluster map graph as a PNG image, HTML, or XML image map (to publish the map on a web-page).
To export cluster map, select "File → Export Cluster Map" menu item. In the wizard dialog, select desired export format and enter an exported file name.