Research Summaries

Back Using Open Source Natural Language Processing to Inform Clearance Worthiness

Fiscal Year 2020
Division Graduate School of Operational & Information Sciences
Department Data Science and Analytics Group
Investigator(s) Dell, Robert F.
Sponsor Army Analytics Group (Army)
Summary The Army Analytics Group and the Defense Vetting Directorate require tools to augment and automate the adjudication process to improve the quality and timeliness of the adjudication of security clearances. This process is largely manual at this time, with adjudicators being required to manually review potentially hundreds of pages of documents to identify relevant information to inform their final decision on clearance worthiness. Ongoing efforts have produced models that provide improving levels of accuracy in identifying the risk but further research is required to explore the potential for the incorporation of information held in text documents associated with each case into the models. In particular, there is a need to understand the potential for the use of open source natural language processing capabilities to retrieve relevant information elements from the large corpus of documents associated with each case and to further reduce this information for inclusion as features in machine learning efforts. A need also exists to conduct an initial comparison of the results of models generated using open source tools with those generated by proprietary systems.
Keywords Software analytics natural language processing
Publications Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal
Data Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal