Hybrid knowledge-based systems

 

A while ago, in grad school, I co-authored my first book on AI. Even back then, there was an understanding that purely data-driven systems will need to work with knowledge bases representing the complexity that may or may not be there in data sets. After all, if the expert is available or the knowledge bases are available, then hybrid system architecture is the way to go. This book was one of the first reports on how a knowledge-based and a machine learning component can be put together in a single system.