Fundamentals of machine learning for analytical forecasting. Algorithms, working examples

Machine learning is often used to build predictive models by extracting patterns from large datasets. These models are used in data forecasting applications, including price forecasting, risk assessment, customer behavior forecasting, and document classification. This introductory textbook offers a detailed and focused examination of the most important approaches to computer learning used in data mining, covering both theoretical concepts and practical applications. The formal mathematical material is supplemented with explanatory examples, and research examples illustrate the application of these models in a broader business context. After discussing the transition from data preparation to understanding the solution, the book describes four approaches to computer learning: information learning, similarity-based learning, probabilistic learning, and error-based learning. The description of each of these approaches is preceded by an explanation of the underlying concept, followed by mathematical models and algorithms illustrated with detailed working examples. Finally, the book discusses methods for evaluating forecasting models and offers two case studies that describe specific data analysis projects at each stage of development, starting from the formulation of a business task and ending with the implementation of an analytical solution. The book is the result of many years of work by the authors in the field of machine learning and data mining and is suitable for use by students in computer science, engineering, mathematics or statistics, graduate students specializing in fields related to data mining, as well as professionals as a reference.