Discovering Message Templates on Large Scale Bitcoin Abuse Reports Using a Two-Fold NLP-Based Clustering Method

【Author】 Choi, Jinho; Lee, Taehwa; Kim, Kwanwoo; Seo, Minjae; Cui, Jian; Shin, Seungwon

【Source】IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS

【影响因子】0.695

【Abstract】Bitcoin is currently a hot issue worldwide, and it is expected to become a new legal tender that replaces the current currency started with El Salvador. Due to the nature of cryptocurrency, however, difficulties in tracking led to the arising of misuses and abuses. Consequently, the pain of innocent victims by exploiting these bitcoins abuse is also increasing. We propose a way to detect new signatures by applying two-fold NLP-based clustering techniques to text data of Bitcoin abuse reports received from actual victims. By clustering the reports of text data, we were able to cluster the message templates as the same campaigns. The new approach using the abuse massage template representing clustering as a signature for identifying abusers is much efficacious.

【Keywords】Bitcoin abuse; text clustering; text template campaign

【发表时间】2022 APR

【收录时间】2022-05-14

【文献类型】理论性文章

【主题类别】

区块链治理-市场治理-数字货币