Various algorithms that allow a computer to learn from data in a way analogous to, but typically very different from, human or animal learning. This can be used as a research tool, or as part of some form of intelligent interface. Machine learning (ML) algorithms have existed for many years, but the vast amounts of data now available allow us to study phenomena in a way that would have been impossible only a few years ago. However, the analysis of such data using machine learning techniques can lead to unintentional cherry picking, as the ML will pick out the few patterns that work amongst a vast number that do not.
Used on pages 78, 92, 142, 143