Unsupervised Social Event Detection via Hybrid Graph Contrastive Learning and Reinforced Incremental Clustering
arXiv:2312.08374 · doi:10.1016/j.knosys.2023.111225
Abstract
Detecting events from social media data streams is gradually attracting researchers. The innate challenge for detecting events is to extract discriminative information from social media data thereby assigning the data into different events. Due to the excessive diversity and high updating frequency of social data, using supervised approaches to detect events from social messages is hardly achieved. To this end, recent works explore learning discriminative information from social messages by leveraging graph contrastive learning (GCL) and embedding clustering in an unsupervised manner. However, two intrinsic issues exist in benchmark methods: conventional GCL can only roughly explore partial attributes, thereby insufficiently learning the discriminative information of social messages; for benchmark methods, the learned embeddings are clustered in the latent space by taking advantage of certain specific prior knowledge, which conflicts with the principle of unsupervised learning paradigm. In this paper, we propose a novel unsupervised social media event detection method via hybrid graph contrastive learning and reinforced incremental clustering (HCRC), which uses hybrid graph contrastive learning to comprehensively learn semantic and structural discriminative information from social messages and reinforced incremental clustering to perform efficient clustering in a solidly unsupervised manner. We conduct comprehensive experiments to evaluate HCRC on the Twitter and Maven datasets. The experimental results demonstrate that our approach yields consistent significant performance boosts. In traditional incremental setting, semi-supervised incremental setting and solidly unsupervised setting, the model performance has achieved maximum improvements of 53%, 45%, and 37%, respectively.
arXiv commentAccepted by Knowledge-Based Systems
References in corpus (12)
- Efficient Estimation of Word Representations in Vector Space
- Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
- Graph Contrastive Learning with Augmentations
- Beyond Product Quantization: Deep Progressive Quantization for Image Retrieval
- Deep Recurrent Quantization for Generating Sequential Binary Codes
- Deep Graph Contrastive Representation Learning
- SimGRACE: A Simple Framework for Graph Contrastive Learning without Data Augmentation
- Graph Contrastive Learning Automated
- TREND: TempoRal Event and Node Dynamics for Graph Representation Learning
- ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning
- Performance Comparison of Incremental K-means and Incremental DBSCAN Algorithms
- AutoGCL: Automated Graph Contrastive Learning via Learnable View Generators