Research Article
No Pie in the Sky: The Digital Currency Fraud Website Detection
@INPROCEEDINGS{10.1007/978-3-031-06365-7_11, author={Haoran Ou and Yongyan Guo and Chaoyi Huang and Zhiying Zhao and Wenbo Guo and Yong Fang and Cheng Huang}, title={No Pie in the Sky: The Digital Currency Fraud Website Detection}, proceedings={Digital Forensics and Cyber Crime. 12th EAI International Conference, ICDF2C 2021, Virtual Event, Singapore, December 6-9, 2021, Proceedings}, proceedings_a={ICDF2C}, year={2022}, month={6}, keywords={Blockchain Digital currency fraud website Ponzi scheme Phishing}, doi={10.1007/978-3-031-06365-7_11} }- Haoran Ou
Yongyan Guo
Chaoyi Huang
Zhiying Zhao
Wenbo Guo
Yong Fang
Cheng Huang
Year: 2022
No Pie in the Sky: The Digital Currency Fraud Website Detection
ICDF2C
Springer
DOI: 10.1007/978-3-031-06365-7_11
Abstract
In recent years, digital currencies based on blockchain technology are growing rapidly. Therefore, many criminal cases related to digital currency also took place. One of the most common ways is to induce victims to invest. As a result criminals can obtain a large number of profits through fraud. Cybercriminals usually design the layout of digital currency fraud websites to be similar to normal digital currency websites. Use some words related to blockchain, digital currency, and project white papers to confuse victims to invest. Once the victims have invested a lot of money, they cannot use digital currency to cash out. Digital currency is also difficult to track due to its anonymity. In this paper, we classified and summarized the existing methods of identifying digital currency scams. At the same time, we collected 2,489 domain names of fraudulent websites in the digital currency ecosystem and conducted statistical analysis from the four aspects of website text, domain names, rankings, and digital currency transaction information. We proposed a method to detect the website based on domain name registration time, website ranking, digital currency exchange rate, and other characteristics. We use the random forest algorithm as a classifier. The experimental results show that the proposed detection system can achieve an accuracy of 0.97 and a recall rate of 0.95. Finally, the case study results show that the system gets better detection and accuracy than other security products.
