Network Analysis with Graph DBs

Over the past decade, ASR has collaborated with federal and state-level agencies to develop a suite of tools for data integration, fraud detection, investigative research, and social network analysis—based entirely on open-source technology. To date, ASR’s analytic solutions have identified and prevented over $500 million in suspicious claims and payments across multiple layers of government.

We would welcome the opportunity to share our thoughts and experiences with you in a one-hour webmeeting. During this briefing, we would provide a demonstration of network analysis and entity resolution tools ASR has developed for three specific use cases:

  • Graph Exploration allows users to explore the nodes and edges within a network analysis graph, view attributes, and iteratively expand the graph to explore surrounding entities and individuals. At the Internal Revenue Service (IRS), ASR supports a Graph Explorer graphical user interface (GUI) with nearly 10,000 distinct users of the application, mostly IRS investigative analysts serving enforcement functions. New graphs are continuously being developed for expanded use prior to and during IRS investigations.
  • Pattern Identification is GUI that allows users to draw a pattern of interest, search the graph for all instances of that pattern, and filter-rank the resulting cases. Pattern Identification enables graph analysts to quickly assess the scope and scale of known patterns of noncompliance (e.g., non-reciprocal reporting of employment relationships).
  • Fraud Detection models are machine learning (ML) and artificial intelligence (AI)-based algorithms (e.g., neural networks, Beta skeleton graphs, conditional random fields) that identify networks of fraud or noncompliance not evident to human analysts. At the IRS, ASR has supported the development and operation of these advanced fraud detection methods to identify crime networks of fraudulent returns missed by traditional methods. These Fraud Detection models reduce the number of false positive selections, increasing the efficiency and effectiveness of over-burdened investigative case analysts.

With this set of graph analytics and visualization tools, ASR can specify patterns in user interfaces, search and graph patterns for further exploration, visualize and filter results, and present filtered graphs intelligence reports.

Download our capabilities statement and indicate your interest in scheduling a briefing and information exchange. We welcome the opportunity to demonstrate our capabilities to you and your OIG colleagues.

Graph DBs for OIGs

Click on the image below to learn more

Network Analysis for Investigators and Compliance Analysts

Click on the image below to learn more