Features

• available for many different platforms
• automatically uses all available keywords for search
• can connect to AutoFocus Server(s) for central administration
• increasing number of plug-ins for support of different dataformats
• more efficient search with suggested search terms
• free, no registration required
• no information about your search behaviour leaves your computer
• no supplier lock-in through use of Open Source technology and Open Standards

Detail: Cluster Maps explained

AutoFocus is a combination of three mechanisms. First, full-text search in AutoFocus gives you the possibility to find information that contains the word or phrase that you entered. Second, metadata facets provide a way to explore the contents of your information sources. Third, Cluster Maps that enable you to get an even better overview of your search results.

Showing a collection as a cluster In this Cluster Map you see a collection of information objects. Documents are shown as yellow spheres, e-mails as blue ones. If this cluster would contain web pages, you would see green spheres. The objects in this cluster share the fact that they all contain the search term biology.
Showing overlap between clusters After adding another search term, in this case taxonomy, the map displays three clusters. The cluster in tokhe middle shows all objects that have both terms. Note that the map also shows two clusters of objects that contain one of the terms. The size of clusters gives you visual feedback on the number of object in a cluster. When the number of objects exceeds 250, the cluster will not show individual objects anymore. Instead a disc with the number of objects in it is shown. When you click on a cluster, AutoFocus will show the objects it contains in a list. See the screenshots section for examples.
Adding terms means more clusters Adding a third term (visualization) results in a more interesting map. The middle cluster is connected to all three search terms. But the map also shows all other combinations of answers too. For example, the intersection of the first two terms in combination with the negation of the third. This is a difference with search engines that return a ranked list of search results. If you are lucky you will find your answer as the first entry of the list. But if you are not, the list won't give you information about how the entries relate to the search terms you used. The Cluster Map will give you that information. Adding more search terms to the map, will result in more complex maps. AutoFocus provides different colour schemes that help you to understand the map. For example, a colour scheme that shows the most connected cluster in a different colour.

Technology

AutoFocus is based on the following technologies:

• Aperture: for extracting and querying full text content and metadata of file systems, mailboxes and web sites.
• Sesame: for fast and scalable storage and querying of metadata.
• Cluster Map Library: for showing Cluster Maps that explain the search results.

Requirements

Operating systems

AutoFocus is supported when it is installed in one of the following environments:

• Windows 98 (Second Edition)
• Windows ME
• Windows NT 4.0 SP6 and later
• Windows 2000 SP3 and later
• Windows XP SP1 and later
• Solaris 8 and later
• Red Hat Linux 8 and later
• SuSe Linux 8 and later
• Mandrake Linux 9 and later
• MacOS X

Note that in principal AutoFocus can run on any platform with a Java virtual machine by using the Java installer, although we cannot officially guarantee correct behaviour on these platforms. We recommend that you use Java SE version 5.0 Update 7. You can download a JSE installer from Sun's Java site.

Hardware requirements

• CPU: the absolute minimum is a Pentium II at 400 MHz, a Pentium III at 1 GHz or better is recommended.
• main memory: minimally 128 MB, 256 MB is recommended.
• disk space requirements: 100 MB + 2 MB per 1000 scanned items.