Detect Camouflaged Spam Content via StoneSkipping: Graph and Text Joint Embedding for Chinese Character Variation Representation

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 6187–6196, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 6187 Detect Camouflaged Spam Content via StoneSkipping: Graph and Text Joint Embedding for Chinese Character Variation Representation Zhuoren Jiang1∗, Zhe Gao2∗, Guoxiu He3, Yangyang Kang2, Changlong Sun2, Qiong Zhang2, Luo Si2, Xiaozhong Liu4† 1 School of Data and Computer Science, Sun Yat-sen University, Guangzhou, China 2Alibaba Group, Hangzhou & Sunnyvale & Seattle, China & USA 3School of Information Management, Wuhan University, Wuhan, China 4School of Informatics, Computing and Engineering, IUB, Bloomington, USA jiangzhr3@mail.sysu.edu.cn, {gaozhe.gz,yangyang.kangyy,qz.zhang,luo.si}@alibaba-inc.com, guoxiu.he@whu.edu.cn, changlong.scl@taobao.com, liu237@indiana.edu Abstract The task of Chinese text spam detection is very challenging due to both glyph and pho- netic variations of Chinese characters. This paper proposes a novel framework to jointly model Chinese variational, semantic, and con- textualized representations for Chinese text spam detection task. In particular, a Variation Family-enhanced Graph Embedding (VFGE) algorithm is designed based on a Chinese char- acter variation graph. The VFGE can learn both the graph embeddings of the Chinese characters (local) and the latent variation fam- ilies (global). Furthermore, an enhanced bidi- rectional language model, with a combina- tion 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. 1 Introduction Chinese orchestrates over tens of thousands of characters by utilizing their morphological in- formation, e.g., pictograms, simple/compound ideograms, and phono-semantic compounds (Nor- man, 1988). Different characters, however, may share the similar glyph and phonetic “root”. For instance, from glyph perspective, character “裸 (naked)” looks like “课 (course)” (homographs), while from phonetic viewpoint, it shares the similar pronunciation with “锣 (gong)” (homo- phones). The form of variations can also be compounded, for instance, “账 (account)”and “帐 ∗These two authors contributed equally to this research. †Corresponding author Figure 1: Character Variations in Chinese Spam Texts (the pinyin codes provide phonetic information and the stroke and Zhengma codes provide glyph information). (curtain)” have the similar structure and pronunci- ation (homonyms). Unfortunately, in the context of spam detection, as shown in Figure 1, spam- mers are able to take advantage of these variations to escape from the detection algorithms (Jindal and Liu, 2007). For instance, in the e-commerce ecosystem, variation-based Chinese spam muta- tions thrive to spread illegal, misleading, and harmful information1. In this study, we propose a novel problem - Chinese Spam Variation Detec- tion (CSVD), a.k.a. investigating an effective Chi- nese character embedding model to assist the clas- sification models to detect the variations of Chi- nese spam text, which needs to address the follow- ing key challenges. Diversity: the variation patterns of Chinese characters can be complex and subtle, which are difficult to generalize and detect. For instance, in the experimental dataset, one Chinese char- acter …

Content truncated to 500 words