Smart De-Identification of Social Media Data

Abstract

Final Technical and Financial Report on the development of a prototype de-identification tool for Twitter. The objective of this work was to develop an understanding of the full range of requirements for a de-identification tool to remove personally identifiable information from social media data, both the structured and the un-structured components, to design such a tool and develop a prototype. The resulting de-identifier is scalable, supports multiple levels of de-identification, and enables smart-de-identification (and so retention of object class information) in a fashion that is under control of the user, supports over-time and cross-data comparison, and meets legal requirements.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 13, 2014
Accession Number
ADA608548

Entities

People

  • Jonathan Storrick
  • Kathleen Carley
  • L. R. Carley

Tags

DTIC Thesaurus Topics

  • Big Data
  • Classification
  • Contracts
  • Crisis Management
  • Data Processing
  • Department Of Defense
  • Electronic Mail
  • Human Trafficking
  • Identification
  • Identification Systems
  • Instructions
  • Media
  • Networks
  • Online Communications
  • Social Media
  • Social Networking Services
  • Social Networks

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Database Systems and Applications