Automated/assisted target behavior model development

OUSD (R&E) MODERNIZATION PRIORITY: Autonomy; Artificial Intelligence/Machine Learning TECHNOLOGY AREA(S): Information Systems The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with the Announcement. Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws. Please direct questions to the Air Force SBIR/STTR HelpDesk: usaf.team@afsbirsttr.us. OBJECTIVE: Identify sources of intelligence reports describing behaviors of targets of interest and methods to extract the information in the reports into a machine readable format. Identify sensor data relevant to the targets of interest that can be extracted from multi-domain sources. The autonomous algorithms use the extracted data to develop adversary patterns of life for various activities. These can be temporal patterns or functional patterns. The algorithms would interpret the patterns and identify a workflow and patterns of operation. These automated extracted patterns would be converted into a format ingestible by Insight and other predictive analytics tools used to augment and potentially replace expert defined patterns. These patterns would be used to support predictive analytics and identify which collections are needed to increase confidence and confirm patterns the enemy is using. In addition, these patterns may need to support flexibility and adapt to variations that change over time. DESCRIPTION: Develop tools that can extract relevant target behavior information from textual and machine readable reports. Using the extracted information automate or assist the analyst in the development Insight readable models of target behaviors, and link at least three different types of sensors and/or from 3 different sensors across 2 different sensing methods from 2 different operating domains to measure indications of the target behavior. The algorithms would produce either a functional or temporal pattern of life that would drive automated patterns of life implemented in Insight or other predictive analytics software tools. PHASE I: Identify an initial domain and target behavior of interest. Identify reports describing the behaviors of target of interest. Identify a data set with sensing information containing the target and behaviors of interest. Prototype capabilities to demonstrate all of the required functions to extract the information from reports and create simple target behavior models for Insight. PHASE II: Refine and integrate prototype capabilities developed in Phase 1. Demonstrate integrated tool for at least three different types of sensors and/or from 3 different sensors across 2 different sensing methods from 2 different operating domains to measure indications of the target behavior. PHASE III DUAL USE APPLICATIONS: Potential to provide commercial functionality to multiple organizations across the Department of Defense for internal and external applications, civic and commercial applciations for automated workflow applications linking information capture systems to define meaning and process steps. This provides the potential to accelerate the process development, reducing manpower requirements, while improving overall quality control. REFERENCES: Machine Assisted Script Curation, Proceedings of NAACL-HLT 2021: Demonstrations, pages 8 17June 6 11, 2021. 2021 Association for Computational Linguistics; Rahul, S. Adhikari and Monika, "NLP based Machine Learning Approaches for Text Summarization," 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC), 2020, pp. 535-538, doi: 10.1109/ICCMC48092.2020.ICCMC-00099; NLP Driven Ensemble Based Automatic SubtitleGeneration and Semantic Video SummarizationTechniqueAswin VB, Mohammed Javed, Parag Parihar, Aswanth K, Druval CR, Anpam Dagar,Aravinda CV1Indian Institute Of Information Technology Allahabad, Prayagraj, Uttar Pradesh, http://arxiv.org/abs/1904.09740 KEYWORDS: Natural language processing; Course of action modeling; Target behavior modeling; Time sequence modeling; indications and warnings; DARPA Insight; Automated model generation; Assisted model generation;