Predicting the structural evolution of networks by applying multivariate time series

• We propose a method for link prediction in time-varying networks.
• We combined time series methods with traditional indexes to improve prediction accuracy.
• We discuss the importance of temporal information and topological information for dynamic network analysis.

In practice, complex systems often change over time, and the temporal characteristics of a complex network make their behavior difficult to predict. Traditional link prediction methods based on structural similarity are good for mining underlying information from static networks, but do not always capture the temporal relevance of dynamic networks. However, time series analysis is an effective tool for examining dynamic evolution. In this paper, we combine link prediction with multivariate time series analysis to describe the structural evolution of dynamic networks using both temporal information and structure information. An empirical analysis demonstrates the effectiveness of our method in predicting undiscovered linkages in two classic networks.