Abstract
Static, point-in-time transaction validation methods may be susceptible to in-session fraud, such as account takeovers, and can impose friction on legitimate users. To support continuous transaction authorization, systems and methods can analyze tokenized behavioral data. On a user device (e.g., a smartphone, smart watch, or wearable device), hardware sensors can capture behavioral data that is converted into privacy-preserving tokens before transmission to a server. A server-side infrastructure may use these tokens to build a real-time behavioral model for a session. During a transaction, this model can be compared to a user's historical baseline to compute a dynamic confidence score. This score may inform a multi-tiered response, ranging from streamlined approval with an injected cryptographic proof for high scores to a proportional, context-appropriate user challenge or a transaction block for low scores. This process may mitigate certain types of fraud while potentially reducing security friction.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Moguluru, Sree Sitamahalakshmi, "Dynamic Transaction Authorization Using Continuous Analysis of Tokenized Behavioral Data", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/11009