Using Textual Analysis to Detect Initial Coin Offering Frauds

Abstract

Initial coin offering (ICO) has attracted a lot of attention from the public in recent years due to its association with potentially fraudulent activities. In order to offer practical implications to investors and regulators when evaluating ICO projects, this study examines the use of textual analysis in detecting potential ICO fraud cases. By using Linguistic Inquiry and Word Count (LIWC), we extracted the textual characteristics of 1,402 English whitepapers that may have been indicators of potential fraud based on the prior literature, including first-person plural pronouns, adverbs, and certainty, and formed a risk index for potentially problematic ICOs. Our findings suggest that the use of these words reflects the warning signals raised by the Securities and Exchange Commission (SEC) about potentially problematic ICO projects, which can therefore be used by regulators and investors when evaluating ICOs. Implications are discussed.