CERIAS Weekly Security Seminar – Purdue University
Mercan Topkara, Hiding the Message Behind the Words: Advances in Natural Language Watermarking
The Internet has become one of the main sources of knowledgeacquisition, harboring resources such as online newspapers, webportals for scientific documents, personal blogs, encyclopedias, andadvertisements. It has become a part of our daily life to search andaccess this immense amount of online information, and more recently wehave also started to contribute to this pool of information our owncreativity in the form of text, images and video. Unfortunately, it isstill an open question as to how we, as authors, can control the waythat the information we create is distributed or re-used.Rights management problems are serious for text since it is much easyfor other people to download and manipulate copyrighted text fromInternet and later re-use it free from control. There is a need for arights protection system that ``travels with the content''. Digitalwatermarking is an information hiding mechanism that embeds thecopyright information in the document. Besides traveling with thecontent of the documents, digital watermarks are also imperceptible(i.e., seamless) to the user, which makes the process of removing themfrom the document challenging.Using linguistic features for information hiding into natural language text is an exciting and new idea. This talk begins with a short surveyof existing technologies in natural language watermarking, and thenfocuses on a recently developed natural language watermarking systemthat is practical, easy-to-use and provides resilience to attacks throughthe use of ambiguity in natural language. The talk is aimed for a generalaudience, and will be self-contained covering the necessary backgroundinformation. About the speaker: Mercan Topkara is a PhD candidate at the Computer Science Departmentof Purdue University working with Mikhail J. Atallah and CristinaNita-Rotaru. She got her Bachelor of Science degree from ComputerEngineering and Information Science Department of Bilkent Universityin 2000. She started her graduate studies at Purdue University inAugust 2001. Her PhD thesis is focused on designing, building andevaluating natural language watermarking systems. Her researchinterests are within the areas of digital watermarking, statisticalnatural language processing, usable security and machine learning. Shehas previously worked as a research intern at AT&T Research Labs, IBMT. J. Watson Research, and Google Research. More information can befound at http://www.cs.purdue.edu/homes/mkarahan.
Senaste avsnitt
En liten tjänst av I'm With Friends. Finns även på engelska.