UMass Database and Information Management Lab

Print this page

Private Dissemination of Communication Traces

Communication traces are detailed records of communication between two entities in a network. These traces arise in a variety of settings and include computer network traces, phone toll records, instant messaging transcripts, among others. Sharing of communication traces across institutions is critical for network research into traffic analysis, communication protocols, routing and security of communication networks. But the public release of these traces remains highly constrained by privacy and security concerns. It is because the traces contain highly sensitive information about the users of the network as well as the network-sensitive information that could be used in attack against the network.

The goal of this work is to understand the threats involved in publication of communication traces and develop a framework to support their safe dissemination. It is required that the published communication traces must have high utility for the network researchers. The publications below address the threats to privacy of the users in the communication traces and propose framework that guarantees absolute utility to the network researchers while protecting the privacy of the users.

Publications

• A Framework for Safely Publishing Communication Traces
  Abhinav Parate and Gerome Miklau
  Conference on Information and Knowledge Management (CIKM) 2009
• A Framework for Safely Publishing Communication Traces
  Abhinav Parate and Gerome Miklau
  University of Massachusetts Full Technical Report 2009-040, 2009
• Anonymizing Network Traces Analyzing Privacy in Enterprise Trace Anonymization
  Bruno Ribeiro, Weifeng Chen, Gerome Miklau, Don Towsley
  Network and Distributed System Security Symposium (NDSS) 2008
• A Framework for Utility-Driven Network Trace Anonymization
  Abhinav Parate and Gerome Miklau
  University of Massachusetts Technical Report 2008-038, 2009

Project Members

• Abhinav Parate (graduate student)
• Bruno Ribeiro (graduate student)
• Gerome Miklau (faculty)
• Don Towsley (faculty)

This project is funded in part by NSF Cybertrust grant (CNS-0627642) Preserving utility while ensuring privacy for linked data, 2006-2009.