Information Access

An examination of the structure of the Web leads to the notion that it could more aptly be named the "maze." In designing an online information retrieval system to guide the user through that maze, staff at the Center for Biological Informatics strive to provide the shortest route to the desired information-while at the same time retaining the dynamic character of the Web, which provides for the serendipity of unexpected discovery.

To enable sorting through, identifying, and obtaining the specific information needed by users, information must first be categorized, labeled, and indexed. The complexity of this task is increased by the distributed nature of information resources, which are located across the Web and have no provision for centralized control. Our approach is to include the user when designing information access applications as an essential first step to ensuring that subsequent transactions will be smooth and natural. In improving access to the National Biological Information Infrastructure, for example, CBI investigates and develops applications in such areas as interface design, metadata, the use of metatags on Web sites, employing a structured vocabulary such as the CERES/NBII Thesaurus Partnership Project and ITIS, conducting searches using such tools as BioBot, visualization of search results, and data warehousing and mining. Such applications are reflected in the current functionality of NBII or are future efforts for improving services.