Visual Interfaces for Semantic Information Retrieval and Browsing [chapter]

Katy Börner
2003 Visualizing the Semantic Web  
Today, an unimaginable amount of human knowledge is available in digital libraries, repositories, on the Internet, etc. Computers have become the entry point to a worldwide network of information, services, and other people. However, the Web is extremely unstructured and heterogeneous, making the efficient and effective discovery of relevant data, services, or expertise rather difficult. In sum, we are facing unparalleled challenges and opportunities for universal knowledge collection, access,
more » ... nd management. Currently, various indexing systems, cataloguing systems, and search engines exist, but their ability to retrieve relevant data from diverse, distributed databases with different formats and multimedia data (text, images, audio, video) is very limited. Among the well-known problems for traditional information systems are the vocabulary inconsistency between user queries and data presentation (Bates, 1998) , and the approaches simple keyword-matching usage of exploiting the frequency of co-occurrence of terms. The vision shared by major information retrieval (IR) experts, as well as the developers of the new "Semantic Web", is a network of distributed repositories where multilingual, multimedia documents can be searched within and across differently indexed and structured collections (Schatz and Chen, 1996) . To make this vision come true, technologies are needed that search repositories despite not only variations in protocols and formats, but also variations in content and meaning. Several approaches to semantic information retrieval and browsing have been proposed and are reviewed in Section 7.2. Visual interfaces exploit powerful human visual processing to ease information filtering and browsing. By connecting information to space, they also support browsing, i.e. the exploration of information space in order to become familiar with it and to locate information of interest. Browsing requires a working knowledge of both applied knowledge organization (typically alphabetical or hierarchical) and the means to navigate in it (Chen et al., 1998) . Section 7.3 reviews related research on visual interfaces for digital libraries and information workspaces. Section 7.4 explains the inner workings of the LVis Digital Library Visualizer. It details data analysis and data visualization and presents two prototypical 99
doi:10.1007/978-1-4471-3737-5_7 fatcat:gecf6t4hjnhvxajxz7x43ge2zi