Abstract

In this paper, we take a step towards a principled method of network composition from multi-layer data. We argue that inter-layer dynamics is a essential component of understanding the structure as a whole. Mathematically, we consider the following abstract problem: given multiple layers of network data over a shared vertex set, and additional parameters for inter-layer transitions, construct a (single) weighted network that best integrates the multi-layer dynamics. In this context, we will also study an empirical use case of the composition framework.

Date

January 15, 2026

Authors

Xiaoran Yan, Shang-Hua Teng, Kristina Lerman

Conference

Social, Cultural, and Behavioral Modeling: 10th International Conference, SBP-BRiMS 2017, Washington, DC, USA, July 5-8, 2017, Proceedings 10

Pages

315-321

Publisher

Springer International Publishing