Abstract:Networks are ubiquitous in science and have become a focal point for discussion in everyday life. Formal statistical models for the analysis of network data have emerged as a major topic of interest in diverse areas of study, and most of these involve a form of graphical representation. Probability models on graphs date back to 1959. Along with empirical studies in social psychology and sociology from the 1960s, these early works generated an active network community and a substantial literature in the 1970s. This effort moved into the statistical literature in the late 1970s and 1980s, and the past decade has seen a burgeoning network literature in statistical physics and computer science. The growth of the World Wide Web and the emergence of online networking communities such as Facebook, MySpace, and LinkedIn, and a host of more specialized professional network communities has intensified interest in the study of networks and network data. Our goal in this review is to provide the reader with an entry point to this burgeoning literature. We begin with an overview of the historical development of statistical network modeling and then we introduce a number of examples that have been studied in the network literature. Our subsequent discussion focuses on a number of prominent static and dynamic network models and their interconnections. We emphasize formal model descriptions, and pay special attention to the interpretation of parameters and their estimation. We end with a description of some open problems and challenges for machine learning and statistics.
What problem does this paper attempt to address?
The main problem that the paper "A Survey of Statistical Network Models" attempts to solve is to provide a comprehensive overview of statistical network models, especially in terms of historical development, main modeling methods and their applications in different fields. Specifically, the objectives of the paper include:
1. **Historical Review**: The paper first reviews the historical development of statistical network models, introducing the development process from early research in social psychology and sociology to modern statistical physics and computer science. This helps readers understand the background and foundation of current research.
2. **Model Classification**: The paper classifies the existing statistical network models, mainly into two categories: static models and dynamic models. Static models focus on explaining the observed set of links based on a network snapshot at a single point in time, while dynamic models focus on the mechanisms by which the network changes over time.
3. **Model Description and Parameter Estimation**: The paper describes in detail the forms of various models, and pays special attention to the interpretation of parameters and their estimation methods. For example, the Erdős–Rényi–Gilbert random graph model, the exchangeable graph model, the p1 model, the exponential random graph model (ERGM), the random graph model with a fixed - degree distribution, the block model, the stochastic block model, community detection, the latent space model, etc.
4. **Practical Application Cases**: The paper shows the applications of these models in different fields through examples of multiple real - data sets, such as social network analysis, protein - interaction networks in biology, email communication networks, etc. These examples help readers understand the practical application value of the models.
5. **Future Research Directions**: Finally, the paper points out some open problems and challenges in current statistical network model research, providing directions for future machine - learning and statistical research.
In general, this paper aims to provide readers with a comprehensive perspective to understand the current situation and development trends of statistical network models, as well as their applications in various fields.