Statistical and Machine Learning Approaches for Network Analysis

2.1 Introduction

In real-world systems, several phenomena occur as a result of intricate interactions among several elements. Networks describe the relationships among elements, and are, thus, simple and powerful tools for describing complicated systems. The concept of networks is universal and can be applied to a wide range of fields (mathematics, computer science, economy, sociology, chemistry, biology, etc.). In recent years, considerable data on interaction has been accumulated using several new technologies and high-throughput methods. Thus, networks have become quite important for understanding real-world systems and extracting knowledge of complicated interactions.

In this chapter, we provide an overview of network science. In particular, we discuss an example of real-world networks and network representation in Section 2.2 and also mention classical network modes such as random networks and lattice networks, which are the basis of network analysis. In Section 2.4 and subsequent sections, we discuss the remarkable statistical properties of networks, and explain network measures. In addition, well-known network models are mentioned for providing a clearer understanding of the statistical properties of networks.

Several excellent books on network science have already been published (e.g., Refs. [1–4]). In addition to these books, this chapter aims to facilitate a clearer understanding of network science.