Basics of machine learning for analytical forecasting: algorithms, working examples and thematic studies. KELLEHER J.D., McNami B., d’Arsi A.

machine learning is often used to build prognostic models by extracting templates from large data sets. These models are used in applications for data forecasting, including price forecasting, risk assessment, customer behavior and documents classification. This introductory textbook offers a detailed and targeted consideration of the most important approaches to computer training, used in intellectual analysis of data covering both theoretical concepts and practical applications. Formal mathematical material is complemented by explanatory examples, and examples of research illustrate the use of these models in a wider business of business. After discussing the transition from the preparation of data to understanding the decision in the book, four approaches to computer training are described: information training, similarities, probabilistic training and error -based training. The description of each of these approaches is preceded by an explanation of the fundamental concept, followed by mathematical models and algorithms illustrated by detailed work examples. Finally, the book discusses the methods for evaluating forecasting models and two thematic studies are proposed that describe specific data analysis projects at each development stage, ranging from the formulation of business and the implementation of the analytical solution. The book is the result of many years of work of authors in the field of machine learning and intellectual data analysis and is suitable for students in the field of computer science, engineering, mathematics or statistics, graduate students specializing in areas related to intellectual data analysis, as well as professionals as a reference book.