Date of Award

3-26-2020

Document Type

Thesis

Degree Name

Master of Science in Cyber Operations

Department

Department of Electrical and Computer Engineering

First Advisor

Richard Dill, PhD

Abstract

The USAF and the DoD use traditional schoolhouses to educate and train personnel. The physical aspects of these schoolhouses limit throughput. A method to increase throughput is to shift towards an asynchronous learning environment where students move through content at individually. This research introduces a methodology for transforming a set of unstructured documents into an organized TM students can use to orient themselves in a domain. The research identifies learning paths within the TM to create a directed KSAT. We apply this methodology in four case studies, each an education or training course. Using a graph comparison metric and the topic identification rates for the TMs, we tested a whitelisting algorithm to identify topics with up to 81% accuracy, and leveraged a standalone LDA algorithm and the same LDA algorithm with ConceptNet for topic naming. The research also produced a KSAT for all case studies and two modified KSATs. The research shows that TMs and KSATs can automatically be created with minimal user input. This methodology could help increase throughput in Air Force education and training pipelines.

AFIT Designator

AFIT-ENG-MS-20-M-052

DTIC Accession Number

AD1103023

Download

Share

COinS