[Re] Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks

Romain Haton, Rayane Ait Ali Yahia, Vincent Gauthier and Amel Bouzeghoub

SAMOVAR, Télécom SudParis, Institut Polytechnique de Paris, Palaiseau, France

In this work, the article Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks [1] is evaluated through a replication study. Replication results examine whether the claims made by the authors are valid. Our goal is to replicate the experiments and achieve the same results as the authors. Our code used for the replication is available at the following address: https://github.com/ComplexNetTSP/JODIE.

References

[1] S. Kumar, X. Zhang, and J. Leskovec. “Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks.” In: Proceedings of the 25th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM. 2019.