Streaming Social Event Detection and Evolution Discovery in Heterogeneous Information Networks

TL;DR

This paper introduces a novel framework for real-time social event detection and evolution analysis in heterogeneous information networks, leveraging a new meta-schema, a specialized graph convolutional network, and streaming clustering techniques.

Contribution

It proposes a new event-based meta-schema, a Pairwise Popularity Graph Convolutional Network (PP-GCN), and a streaming detection framework for evolving social events in HINs, advancing real-time social event analysis.

Findings

Framework outperforms existing methods in detection accuracy.

Effective in capturing event evolution over time.

Demonstrated on real-world social media data.

Abstract

Events are happening in real-world and real-time, which can be planned and organized for occasions, such as social gatherings, festival celebrations, influential meetings or sports activities. Social media platforms generate a lot of real-time text information regarding public events with different topics. However, mining social events is challenging because events typically exhibit heterogeneous texture and metadata are often ambiguous. In this paper, we first design a novel event-based meta-schema to characterize the semantic relatedness of social events and then build an event-based heterogeneous information network (HIN) integrating information from external knowledge base. Second, we propose a novel Pairwise Popularity Graph Convolutional Network, named as PP-GCN, based on weighted meta-path instance similarity and textual semantic representation as inputs, to perform fine-grained social event categorization and learn the optimal weights of meta-paths in different tasks. Third, we propose a streaming social event detection and evolution discovery framework for HINs based on meta-path similarity search, historical information about meta-paths, and heterogeneous DBSCAN clustering method. Comprehensive experiments on real-world streaming social text data are conducted to compare various social event detection and evolution discovery algorithms. Experimental results demonstrate that our proposed framework outperforms other alternative social event detection and evolution discovery techniques.