TY - GEN
T1 - DAMe
T2 - 33rd ACM International Conference on Information and Knowledge Management, CIKM 2024
AU - Yu, Xiaoyan
AU - Wei, Yifan
AU - Li, Pu
AU - Zhou, Shuaishuai
AU - Peng, Hao
AU - Sun, Li
AU - Zhu, Liehuang
AU - Yu, Philip S.
N1 - Publisher Copyright:
© 2024 ACM.
PY - 2024/10/21
Y1 - 2024/10/21
N2 - Training social event detection models through federated learning (FedSED) aims to improve participants' performance on the task. However, existing federated learning paradigms are inadequate for achieving FedSED's objective and exhibit limitations in handling the inherent heterogeneity in social data. This paper proposes a personalized federated learning framework with a dual aggregation mechanism for social event detection, namely DAMe. We present a novel local aggregation strategy utilizing Bayesian optimization to incorporate global knowledge while retaining local characteristics. Moreover, we introduce a global aggregation strategy to provide clients with maximum external knowledge of their preferences. In addition, we incorporate a global-local event-centric constraint to prevent local overfitting and ''client-drift''. Experiments within a realistic simulation of a natural federated setting, utilizing six social event datasets spanning six languages and two social media platforms, along with an ablation study, have demonstrated the effectiveness of the proposed framework. Further robustness analyses have shown that DAMe is resistant to injection attacks.
AB - Training social event detection models through federated learning (FedSED) aims to improve participants' performance on the task. However, existing federated learning paradigms are inadequate for achieving FedSED's objective and exhibit limitations in handling the inherent heterogeneity in social data. This paper proposes a personalized federated learning framework with a dual aggregation mechanism for social event detection, namely DAMe. We present a novel local aggregation strategy utilizing Bayesian optimization to incorporate global knowledge while retaining local characteristics. Moreover, we introduce a global aggregation strategy to provide clients with maximum external knowledge of their preferences. In addition, we incorporate a global-local event-centric constraint to prevent local overfitting and ''client-drift''. Experiments within a realistic simulation of a natural federated setting, utilizing six social event datasets spanning six languages and two social media platforms, along with an ablation study, have demonstrated the effectiveness of the proposed framework. Further robustness analyses have shown that DAMe is resistant to injection attacks.
KW - federated learning
KW - model aggregation
KW - social event detection
UR - http://www.scopus.com/inward/record.url?scp=85209996432&partnerID=8YFLogxK
U2 - 10.1145/3627673.3679551
DO - 10.1145/3627673.3679551
M3 - Conference contribution
AN - SCOPUS:85209996432
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 3052
EP - 3062
BT - CIKM 2024 - Proceedings of the 33rd ACM International Conference on Information and Knowledge Management
PB - Association for Computing Machinery
Y2 - 21 October 2024 through 25 October 2024
ER -