Reinforced, incremental and cross-lingual event detection from social messages

Detecting hot social events from social messages is crucial as it highlights significant happenings. However, the challenge is that the existing event detection methods are generally confronted with ambiguous events features, dispersive text contents, and multiple languages. In this paper, we present a novel reinForced, incremental and cross-lingual social Event detection architecture, namely FinEvent, from streaming social messages. Concretely, we first model social messages into heterogeneous graphs. Secondly, we propose a new reinforced weighted multi-relational graph neural network framework to select optimal aggregation thresholds to learn social message embeddings. To solve the long-tail problem, a balanced sampling strategy guided Contrastive Learning mechanism is designed for incremental social message representation learning. Thirdly, a new Deep Reinforcement Learning guided density-based spatial clustering model is designed to select the optimal minimum number of samples and optimal minimum distance between two clusters. Finally, we implement incremental social message representation learning based on knowledge preservation on the graph neural network and achieve the transferring cross-lingual social event detection. We conduct extensive experiments to evaluate the FinEvent on Twitter streams, demonstrating a significant and consistent improvement in model quality with 14%-118%, 8%-170%, and 2%-21% increases in performance on offline, online, and cross-lingual social event detection tasks.