Recent themes in case-based reasoning and knowledge discovery

Case-based reasoning (CBR) systems have tight connections with machine learning and knowledge discovery and often incorporate diverse knowledge discovery functionalities and algorithms. This article presents themes identified in work presented at recent workshops on synergies between CBR and knowledge discovery. Among the main themes appear Big Data, with cases involving signals, images, texts, and other complex types of data; similarity metric discovery, in the form of weight spaces, feature weights, and feature selection; adaptation knowledge; explainability and transparency; and user centeredness and interactivity. Researchers highlight the advantages of case-based reasoning in terms of its lazy learning, explainability, user centeredness, and interactivity when performing knowledge discovery, as well as how diverse knowledge discovery methods can improve CBR.