A Scalable Approach to Modeling Cascading Risk in the MDAP Network

Abstract

The overarching goal of our multi-year research agenda is to proactively model the non-linear cascading effects of interdependencies in Major Defense Acquisition Program (MDAP) networks. We use this to identify the associated data acquisition challenges so that appropriate governance mechanisms can then be isolated. In this paper, we describe our progress towards a scalable, automated approach for extracting and analyzing the data in the form of Selected Acquisition Reports (SAR) and Defense Acquisition Executive Summaries documents of a network of MDAPs to support a decision-theoretic risk prediction model. Automation is necessitated by the volume and complexity of the data. We will discuss the role of topic modeling, image extraction, and identification of topological features of the MDAP network in this approach.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 30, 2014
Accession Number
ADA612937

Entities

People

  • Anita Raja
  • Ansaf Salleb-aouissi
  • Mohammad Hasan
  • Shalini Rajanna

Organizations

  • University of North Carolina at Charlotte

Tags

Communities of Interest

  • Engineered Resilient Systems

DTIC Thesaurus Topics

  • Acquisition
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Automata Theory
  • Computer Languages
  • Computer Science
  • Computer Vision
  • Computers
  • Data Acquisition
  • Electrical Engineering
  • Identification
  • Image Processing
  • Machine Learning
  • Military Acquisition
  • Network Science
  • Robotics
  • Supervised Machine Learning

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Computer Networking
  • Defense Acquisition Program Management