Evolving knowledge graph similarity for supervised learning in complex biomedical domains

摘要

Knowledge discovery in complex domains can be a challenge for data mining methods, which are typically limited to agnostic views of the data, without being able to gain access to its context and meaning. It is widely recognized that the performance of data mining methods can improve significantly when additional relations among the data objects are taken into account, a strategy employed in relational data mining and Inductive Logic Programming [ ].