Portable information filtering system: network information servers

A communications link can only be considered successful when the desired information is delivered to a user of the communications service. In these days of overwhelming digital information broadcast via numerous wide-area networks including the Internet and corporate intranets, the ability of an individual user to obtain relevant, needed information is becoming increasingly difficult. Information filtering systems perform a data extraction function, analyzing an incoming stream of information and selecting information that matches the needs of its users. Inclusion of an information filtering mechanism in the communications system ensures that the desired information is delivered to a user. Currently, information filtering systems are based on proprietary implementations that are not easily moved to other platforms, do not readily scale for larger volumes of information, and cannot be easily modified to handle new forms of data. This paper presents an overview of the work we have done using the relational database model to implement an information filtering system. We describe the prototype information filtering system we have developed and its initial system accuracy and performance measurements. Use of the relational model provides the information filter system with the critical features of portability, scalability, and flexibility