Intelligent content-based document delivery via automatic filtering profile generation

We develop an intelligent document delivery approach for filtering text inf ormation. Our approach can conduct content-based filtering via a machine le arning technique which automatically constructs a filtering profile from tr aining examples. The profiles, encoded in rule representation, are easily u nderstood by human. Good features of high predictive power for the learning process are automatically extracted from the document content. As a result , our approach is able to operate without any prior information or restrict ion of the topic areas and yet achieve the filtering task. We have conducte d an extensive simulation study to analyze the performance of our approach. We have also implemented a practical intelligent news article delivery sys tem based on our approach. Both simulation study as well as practical exper iments use real-world document collections and the results demonstrate that our approach is effective. (C) 1999 John Wiley & Sons, Inc.