Network Models for Data Science

This text on the theory and applications of network science is aimed at beginning graduate students in statistics, data science, computer science, machine learning, and mathematics, as well as advanced students in business, computational biology, physics, social science, and engineering working with large, complex relational data sets. It provides an exciting array of analysis tools, including probability models, graph theory, and computational algorithms, exposing students to ways of thinking about types of data that are different from typical statistical data. Concepts are demonstrated in the context of real applications, such as relationships between financial institutions, between genes or proteins, between neurons in the brain, and between terrorist groups. Methods and models described in detail include random graph models, percolation processes, methods for sampling from huge networks, network partitioning, and community detection. In addition to static networks the book introduces dynamic networks such as epidemics, where time is an important component.

• Presents both theory and applications of network models and addresses computational issues
• Describes a wide range of modern methods for analysing complex networks, including measures for comparing huge networks and methods for analysing social and biological networks that change over time
• Real-world applications from technological networks, information networks, financial networks, and social networks show the concepts and methods in action