The changing science of machine learning

The latest issue of Machine Learning dealt with the changes being introduced in the area. Early research on machine learning adopted an informal approach to evaluation. Kibler and Langley were two researchers who established a framework for such an experimental science of machine learning, including examples from the emerging literature in this area. The experimental effort in the area was aided by another development when David Aha, a PhD student at UCI started collecting data sets for use in empirical studies of machine learning. The early research effort machine learning was also characterized by an emphasis on symbolic representations of learned knowledge, such as production rules, decision trees, and logical formulae. One of the significant changes introduced in the area involved an increased emphasis on classification and regression tasks in comparison with more complex tasks, such as reasoning, problem solving, and language understanding that had played important roles earlier.