ALEXANDRIA – A knowledge platform on the Internet

The city of the same name on Egypt's Mediterranean coast was home to the ancient world’s most famous library. Even then, its precise cataloguing system facilitated access to knowl-edge. Inspired by this model, ALEXANDRIA researchers are working to develop a knowledge platform on the Internet that will use entirely new ways of interacting to facilitate access to knowledge.

The starting point for the ALEXANDRIA application scenario is the growing importance of Internet users in what is known as Web 2.0. In addition to using search machines, e-mail and chat services, users are actively making their own knowledge available on the Internet. Common examples of this new kind of interaction are blogs, the online encyclopedia Wikipedia and social networks like Face-book and Xing. These platforms would be inconceivable without users who provide and exchange knowledge and information.

Collecting knowledge and making it accessible
ALEXANDRIA is seeking a better way of linking the knowledge that is available in Web 2.0 and work-ing to make that knowledge more readily accessible. To that end, a knowledge platform is being de-veloped that not only collects information, but also helps to manage steadily growing quantities of data. This knowledge platform will help users publish, process and search for content. In order to make the platform intuitive and easy to use, researchers are developing semantic technologies and classification systems that use so-called metadata to link even large quantities of data in a logical way.

Comprehensive access to knowledge
To demonstrate the possibilities offered by this type of knowledge platform, ALEXANDRIA is initially focusing on “history and current events” to illustrate how the application works. With the help of an easy-to-navigate user interface, a demonstration shows how easy it is to access knowledge using newly developed technologies.

Instead of using key words to search for information, as we do today, the user can ask the system a question in normal language. For example, if asked which 20th century Olympic athletes were natives of Berlin, the system is able to provide the relevant names as well as additional information about their lives and athletic achievements. By relating various categories of results – such as persons, places or events – to one another, the system is able to understand human knowledge and reproduce it in an organized fashion.

ALEXANDRIA also helps to display knowledge so that it can be easily visualized. Carrying out a search for information on the Olympics, for example, the knowledge platform combines data on past and future summer and winter games with geographical and scheduling information. The user is then able to find the various Olympic sites on a world map, navigate on a time line through the milestones of Olympic history, from the ancient world up to the present day, and access other information such as the winners in specific disciplines.

Integrating users' knowledge
Users determine which categories and types of information should be highlighted, by evaluating the relevance of specific data on a regular basis. They can also provide a detailed assessment of other information that is made available to them.

During the entry process, ALEXANDRIA indicates which portions of a query are automatically recog-nized. This allows the user to make corrections and narrow down the query. For example, if the user wants to know where Neil Armstrong was born, ALEXANDRIA automatically asks whether the Ameri-can astronaut or the Canadian athlete is meant.

Questions that the system is unable to answer automatically are passed on to other users who have the necessary information, and they are asked to contribute relevant facts and multimedia files to the platform, leading to an ongoing expansion of the knowledge base. Such contributions can also be made using natural language. For instance, the system automatically recognizes the sentence “Neil Armstrong was born in 1932” and incorporates that information into the knowledge base.

To make users an integral component of the ALEXANDRIA platform, the system learns which user has what information and to what extent each user is contributing to the knowledge base. This also makes it possible to bring users together, based on their shared interests, to form knowledge communities.

Mobile access to knowledge
ALEXANDRIA is to be made available through mobile devices such as mobile telephones, cameras or GPS equipment, so that users will have access to the platform at any time and place.
When a user looks at an object through the camera of his mobile telephone, ALEXANDRIA automati-cally superimposes semantic content on the picture, taking into account the user's location and the direction in which he is looking. Historical photographs might be displayed on a current photo of a building, for example, so that the user has a visual experience of traveling through time. Additional information and linked content can be accessed simply by touching the screen. The user's experience is therefore no longer limited to "searching," but also involves "browsing" in semantic contexts.