Inductive Databases and Knowledge Scouts

Explore the concept of knowledge scouts in this 1 hour 25 minute lecture by Ryszard S. Michalski from the Center for Language & Speech Processing at Johns Hopkins University. Delve into the world of inductive databases and learn how intelligent agents called knowledge scouts operate within them to automatically search for target knowledge. Discover the Knowledge Generation Language (KGL-1), a high-level query language that integrates various data mining and machine learning programs with standard data and knowledge management operations. Understand how knowledge scouts are guided by user interest models to find strong patterns in data or specific knowledge required by users. Examine the representation of discovered patterns through association rules in attributional calculus and association graphs, which visually depict multi-argument relationships among concepts with relative strength indicators. Gain insights from two experimental knowledge scouts: one exploring relationships between lifestyles, environmental conditions, symptoms, and diseases in a large medical database, and another searching for patterns in children's behavior using the National Youth Survey database. Assess the potential utility of this methodology for various data mining applications based on preliminary results presented in this informative talk.