FreqyWM: Frequency WaterMarking for the New Data Economy
Devriş Isler, PhD student at IMDEA Networks Institute, Madrid, Spain
We present a novel data engineering technique for modulating the appearance frequency of a few tokens within a dataset for encoding an invisible watermark that can be used to protect ownership rights upon data. We develop optimal as well as fast heuristic algorithms for creating and verifying such watermarks. We also demonstrate the robustness of our technique against various attacks and derive analytical bounds for the false positive probability of erroneously “detecting” a watermark on a dataset that does not carry it.
About Devriş Isler
Devriş Isler is a PhD student at Data Transparency Group under the supervision of Prof. Nikolaos Laoutaris. Devriş received his MSc degree from Cryptography, Security, and Privacy Research Group, Koç University, İstanbul. His research interests include applied cryptography, privacy (e.g., data privacy, privacy perceptions) and usable security. Devriş is currently working on data ownership in data economy. He always enjoys solving security and privacy problems by taking advantage of cryptography (e.g., MPC, FHE, ZKP).
This event will be conducted in English