From Closing Triangles to Closing Higher-Order Motifs
Ryan A. Rossi, Anup B. Rao, Sungchul Kim, Eunyee Koh, Nesreen K. Ahmed · Companion Proceedings of the Web Conference 2020 · 2020
This work introduces higher-order ranking and link prediction methods based on closing higher-order network motifs. In particular, we propose the general notion of a motif closure that goes beyond simple triangle closures and demonstrate that these new motif closures often outperform triangle-based methods. This result implies that one should consider other motif closures beyond simple triangles. We also find that the “best” motif closure depends highly on the underlying network and its structural properties. Furthermore, the methods are fast and efficient for real-time applications such as online visitor stitching, web search, and recommendation. The experimental results indicate the importance of these new motif closures. Finally, the new motif closures can serve as a basis for developing better (un)supervised ranking/link prediction methods.