Unsupervised learning for robust Bitcoin fraud detection

摘要: The rampant absorption of Bitcoin as a cryptographic currency, along with rising cybercrime activities, warrants utilization anomaly detection to identify potential fraud. Anomaly plays pivotal role in data mining since most outlying points contain crucial information for further investigation. In the financial world which network is part by default, amounts fraud detection. This paper investigates use trimmed k-means, that capable simultaneous clustering objects and multivariate setup, detect fraudulent activity transactions. proposed approach detects more transactions than similar studies or reports on same dataset.