Evolution characteristics of the network core in the Facebook

Area of Science:

• Social Network Analysis
• Network Science
• Computational Social Science

Background:

• Static network analysis is well-established.
• Dynamic evolution of online social network cores is crucial but less understood.

Purpose of the Study:

• To empirically investigate the evolving characteristics of the Facebook core.
• To analyze the stability, topological properties, node lifespan, and connection patterns within the Facebook core.

Main Methods:

• K-core decomposition applied to 28 time-stamped snapshots of Facebook link (FL) and wall (FW) datasets.
• Analysis of evolving topological properties: k-core value, assortative coefficient, clustering coefficient, and average shortest path length.
• Investigation of node lifespan within the core and core-whole network connection preferences.

Main Results:

• Core sizes for FL and FW networks remained relatively constant (approx. 672 and 373 nodes) despite network expansion.
• Nodes within the core exhibited increasing interconnectedness over time.
• Over 50% of core nodes persisted for over a year, with 19% remaining from the initial snapshot.
• Core nodes preferentially connected to other high k-core value nodes over high-degree nodes.

Conclusions:

• The Facebook core demonstrates remarkable stability in size and node persistence.
• Core nodes exhibit evolving interconnectedness and distinct connection strategies.
• Findings offer insights into the structural dynamics of large-scale online social networks.