Introduction

Abstract

After a brief description of the pattern recognition and machine learning areas, this chapter sets down basic mathematical concepts and notation used throughout the book. It introduces the key notions of prediction and prediction error for supervised learning. Classification and regression are introduced as the main representatives of supervised learning, while PCA and clustering are mentioned as examples of unsupervised learning. Classical complexity trade-offs and components of supervised learning are discussed. The chapter includes examples of application of classification to Bioinformatics and Materials Informatics problems.