Dynamic Embeddings for User Profiling in Twitter

Abstract
In this paper, we study the problem of dynamic user profiling in Twitter. We address the problem by proposing a dynamic user and word embedding model (DUWE), a scalable black-box variational inference algorithm, and a streaming keyword diversification model (SKDM). DUWE dynamically tracks the semantic representations of users and words over time and models their embeddings in the same space so that their similarities can be effectively measured. Our inference algorithm works with a convex objective function that ensures the robustness of the learnt embeddings. SKDM aims at retrieving top-K relevant and diversified keywords to profile users' dynamic interests. Experiments on a Twitter dataset demonstrate that our proposed embedding algorithms outperform state-of-the-art non-dynamic and dynamic embedding and topic models.

Citation
Liang S, Zhang X, Ren Z, Kanoulas E (2018) Dynamic Embeddings for User Profiling in Twitter. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD ’18. Available: http://dx.doi.org/10.1145/3219819.3220043.

Acknowledgements
This work was supported by the King Abdullah University of Science and Technology (KAUST), Saudi Arabia.

Publisher
Association for Computing Machinery (ACM)

Journal
Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD '18

Conference/Event Name
24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018

DOI
10.1145/3219819.3220043

Additional Links
https://dl.acm.org/citation.cfm?doid=3219819.3220043

Permanent link to this record