Hearts on the Blockchain: Latent Class Analysis of Transaction Pathways, Risk Indicators, and Money Laundering in Cryptocurrency-Facilitated Romance Scams

Abstract

Cryptocurrency has become a dominant payment mechanism in contemporary romance scams, yet victims’ transaction experiences are not uniform and may follow distinct pathways with different risk profiles and laundering consequences. Using 106 U.S.-based cryptocurrency-facilitated romance scam complaints drawn from Chainabuse and validated through open-source blockchain forensics, this study applies latent class analysis to identify subgroups of victims based on transaction modalities and to examine how risk indicators relate to monetary loss and downstream money-laundering techniques. Results support a two-class solution: (1) fake investment service websites, where offenders route deposits through platform-like dashboards with fabricated balances, and (2) direct deposit crypto wallets, where victims transfer cryptocurrency directly to offender-controlled addresses. Within-class heterogeneity was evident: cryptocurrency type (Bitcoin/Ethereum vs. Tron), deception type (investment vs. emergency), and blockchain risk flags showed differential associations with losses across transaction contexts. Distal outcome analyses indicated that laundering strategies varied by class: direct deposit cases exhibited heavier reliance on anonymization (mixers/tumblers), whereas fake investment website cases more frequently involved self-funding and swaps. Framed by Cyber Routine Activities Theory, findings highlight how routine digital behaviors and weak digital guardianship shape victim pathways and offender concealment strategies, informing targeted prevention, compliance screening, and investigative prioritization in cryptocurrency-enabled romance fraud.