An Information Brokering Framework

An Information Brokering Framework,Andreas Ludwig,Ulrich Guntzer

An Information Brokering Framework   (Citations: 1)
BibTex | RIS | RefWorks Download
The Internet and the WWW in particular offer a huge and constantly growing amount of information. Most of the information on the WWW is contained in Web pages that are dynamically prepared in response to some sort of user query (constituting the so-called "Hidden" or "Deep" Web). Other information services are not Web-based but can still be accessed via the Internet, e.g. services which are based on standards like Z39.50 or LDAP. All this infor- mation is beyond the reach of the popular search engines. In this paper, we describe the architecture of an In- formation Broker. The Information Broker is an intelli- gent agent which provides homogeneous access to hetero- geneous information services on the Internet. Sources of this heterogeneity are access protocols, query facilities and capabilities and data formats. We merely require the infor- mation services to be accessible over the Internet and to offer a structured query facility which is based on cate- gory/value pairs. The information provided by the infor- mation services can be of diverse domains. The heterogeneous information services are incorpo- rated into the Information Broker by the use of wrappers. All information services which are based on a standard like Z39.50 or LDAP can be incorporated by customizing a sin- gle generic wrapper, for other information services indi- vidual specific wrappers are needed. All information handling within the Information Broker is XML-based. We introduce the so-called Common Query Markup Language which is used as an interface for the propagation of the user's query to the wrappers. The Information Broker controls the selection of ap- propriate information services, their (parallel) invocation through the wrappers, the combination and postprocessing of the returned query results and facilitates the definition of complex information gathering tasks and strategies.
Cumulative Annual
    • ...We think that PROLOG based Web services can be valuable components of many distributed information systems, for example in an information broker ([7]) to support complex information gathering and integration strategies or to control a mulit‐agent system ([5])...

    Bernd D. Heumesseret al. Web Services Based on Prolog and Xml

Sort by: