The Use of Analytic Decision Game (ADG) Methods for Test and Evaluation of Hard and Soft Data Fusion Systems and Education of a New Generation of Data Fusion Analysts
Abstract
The study and practice of human reasoning, has undergone significant change in recent years from both a theoretical and methodological perspective. With the advent of ubiquitous communication, advances in computing and numerous available analytic support tools, the push for instantaneous reporting and analysis has become the expectation. This is particularly prevalent in the intelligence community (IC) and in the military, where commanders (and their leadership) expect not only a bird's-eye view of operations as they happen, but a play-by-play analysis of operational effectiveness. While we have seen rapid technological advances to support collections analysis and dissemination; it is unclear if our human analysts made similar advances or whether they have even benefitted from the technological advances. Are human analysts capable of making instantaneous assessments or are they simply restating what the sensor/computing "black-box" is spitting out? This paper explores the use of the analytic decision game (ADG) as a method to train, not only the human analyst, but the data fusion systems that inform much of the black-box technology being touted as the next intelligence solution. The ADG is an adaptation of a proven training and planning methodology - the military war game. It presents students with a set of realistic problems and challenges them to work toward actionable intelligence, course of action or probable solution. Such training enhances the capability of the human-in-the-loop in a data fusion system (viz., the "human fuser") as well as providing a basis for training and testing on the same data sets. The ADG can also be used as a database for the test and evaluation of algorithms and the data fusion systems they support. The ADG is not the solution; but the pathway. This paper describes the concept of ADG and describes both research and training being conducted at Penn State using new multi-source synthetic data sets.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 2012
- Accession Number
- ADA581144
Entities
People
- David L. Hall
- Jacob L. Graham
Organizations
- Pennsylvania State University