Detect Camouflaged Spam Content via StoneSkipping: Graph and Text Joint Embedding for Chinese Character Variation Representation
This work addresses spam detection for Chinese text users, offering an incremental improvement by integrating graph and text embeddings to handle character variations.
The paper tackled the challenge of detecting Chinese text spam by addressing glyph and phonetic character variations, proposing a joint graph and text embedding framework that outperformed state-of-the-art models on SMS and review datasets.
The task of Chinese text spam detection is very challenging due to both glyph and phonetic variations of Chinese characters. This paper proposes a novel framework to jointly model Chinese variational, semantic, and contextualized representations for Chinese text spam detection task. In particular, a Variation Family-enhanced Graph Embedding (VFGE) algorithm is designed based on a Chinese character variation graph. The VFGE can learn both the graph embeddings of the Chinese characters (local) and the latent variation families (global). Furthermore, an enhanced bidirectional language model, with a combination gate function and an aggregation learning function, is proposed to integrate the graph and text information while capturing the sequential information. Extensive experiments have been conducted on both SMS and review datasets, to show the proposed method outperforms a series of state-of-the-art models for Chinese spam detection.