Learning in Big Data: Introduction to Machine Learning

Abstract

Abstract Machine learning (ML) is a subset of artificial intelligence that develops dynamic algorithms capable of data-driven decisions, in contrast to models that follow static programming instructions. ML is concerned with enabling computer programs automatically to improve their performance at some tasks through experience. Astronomy and geosciences are two areas where the application of ML can be very fruitful. While the adoption of ML methods in astronomy and geosciences has been slow, there are several published studies using ML in these disciplines. This chapter introduces and evaluates several ML techniques. Special attention is given to inductive learning, which is among the most mature of the ML approaches currently available. The supervised, unsupervised, semisupervised and reinforcement learning types are described. ML algorithms are programs of data-driven inference tools that offer an automated means of recognizing patterns in high-dimensional data. Current trends and recent developments in ML algorithms are discussed. Scalable ML algorithms and frameworks are also described. Selected case study applications in which ML techniques have been successfully deployed in astronomy and geosciences are described. The chapter concludes with a summary of some of the key research issues in ML related to astronomy and geosciences, with emphasis on the scope for the application of ML algorithms to the rapidly increasing volumes of astronomical and remotely sensed geophysical data for geological mapping and other problems.