Towards Real-Time Program Awareness via Lexical Link Analysis

Abstract

DoD acquisition is an extremely complex system, comprised of myriad stakeholders, processes, people, activities, and organizational structures. We believe that the application of a data-driven automation system namely, Lexical Link Analysis (LLA) can facilitate acquisition researchers data sense-making dilemma and help reveal important connections (concepts) and patterns derived from dynamic, voluminous, and on-going data collection. In this past year, we have demonstrated the LLA method to discover valid associations among disparate, unstructured data sets that would have otherwise required lengthy and expensive man-hours to achieve. We analyzed how Trident Warrior 10 technology capabilities link to classified Navy Urgent Need Statements (UNSs). We validated lexical links against the links identified by human experts in the context of realistic, large-scale data sets. We demonstrated using the LLA methods to discover statistically significant correlations. We discovered that the current congressional budget justification practice for Research, Development, Test and Evaluation (RDT&E) tends to allocate resources to avoid overlapping efforts and to fund new and unique projects. We also discovered that the Program Elements (PEs) that match the warfighters requirements obtain more overall attention and less budget reduction compared to the ones without matches. This effort will result in assisting acquisition professionals in improving their decision-making among competing programs and in selecting those that satisfy Navy objectives, thus achieving the Navys goal of improved operational capability.