|
Why is it called"Silver
Bullet"?
Corporate
Personnel
Recent Contracts
Office Locations
Seaport
Contact Us
|
|
Title
|
Client
|
Description
of Work Performed
|
|
Cyber Ontology (CybOnt) Data Fusion
|
Intelligence and Information Warfare Directorate (I2WD), US
Army Combat Capabilities Development Command, (CCDC)
|
CybOnt performs ontology-based fusion for cyber threat
behavior estimation to contribute to an operator's cyber Situational
Awareness (SA) and Situational Understanding (SU). It is unique in that, (1) it is
architected following the Joint Directors of Laboratories (JDL) fusion
levels, (2) it uses formal ontology for the T-Box (types) and A-Box (actuals), and (3) it computes mathematically-principled
-- and thus robust -- likelihood ratios of attack behavior
hypotheses. Inference links are
visualized in a graph database tool that allows customized viewing
tailored to operator requirements.
The likelihood ratios can be thresholded
to give operators control over display clutter. It runs in a tactical cloud environment
and uses big data technologies. As
we continue to develop probabilistic fusion results in the forthcoming
phases, we also plan to incorporate the advances in Deep Learning (DL) in
a hybrid architecture. Architectures we are looking into are:
(a) bootstrapping where the Probabilistic Reasoner (PR) is used to
develop training sets for DL and the DL living learning is used to adjust
probabilistic reasoner’s knowledge base, (b)
using the PR for rapid data triage by processing some subset of the
observations and features to reduce their dimensionality for the DL, (c)
augmenting, where the PR and DL both receive the sensor inputs and PR’s
hypotheses and Λ become additional inputs to the DL, and (d) as a validator and explainer in which the PR is used to
validate the DL has not be spoofed by adversarial AI or come into an
environment it does not know. The
PR explains DL results it agrees with using pedigree or provenance embedded
in fusion file and displayable in PR’s graph displays.
|
|
Systems Engineering and Architecture
|
Space and Naval Warfare Systems Center Atlantic
(SPAWARSYSCEN-LANT)
|
This contract supports enterprise requirements analysis and
assessments, architecture, human systems integration, modeling and
simulation services and future capability engineering in support of
various DoN, DoD and Federal civilian agency
systems.
|
|
Naval Enterprise Architecture Repository (NEAR)
|
Assistant Secretary of the Navy for Research, Development,
and Acquisition (ASN RDA)
|
Silver Bullet developed translators to and from enterprise
architecture tools (IBM System Architect, IBM Rational Software
Architect, No Magic Magic Draw) and the DoD Architecture
Framework (DoDAF) formal ontology, the DoDAF Meta Model (DM2), and its XML Physical Exchange Specification (PES). These translators enable Model Based
Systems Engineering (MBSE) model reuse in conjunction with data exposure
for future architecture and systems engineering analytics, execution, and
simulation. The data exposure
enables cross-architecture analysis for interoperability assessment, cybersecurity risk assessment, mission capability gap
analysis, and portfolio management.
The PES translators were successfully used to transform Joint
Information Environment (JIE) architecture Magic Draw files developed by
the JIE Technical Synchronization Organization (JTSO) to PES for the DoD
CIO Enterprise Architecture and Engineering Panel (EAEP).
|
|
Semantic Distance Function (SDF)
|
IRaD
|
SDF computes semantic distances between pairs of text
documents in the form of likelihood ratios of the likelihood that the
same concepts would have produced the pair compared to the likelihood of
the null hypothesis. What is
unique is, (1) the principled mathematics, (2) the
concept of semantic mass and semantic mass ratios, and (3) an innovative
way to compute the expected semantic mass ratio for the null hypothesis. SDF was inspired by long-standing JDL
Level 2 fusion track correlation problem.
A key element is a Gram Frequency Data Base (GFDB) implemented in Hbase for quick lookup of gram frequencies by
distributed processors computing the semantic distance. For some experiments the GFDB was
initialized using Google's Web 1Terabyte database of one-to-five gram
frequencies. For cyber
experiments, a domain-specific GFDB was developed by processing cyber corpii including Exploit DB (40K entries) and the
Common Vulnerabilities and Exposures (CVE) database (100K entries). The GFDB adapts as input documents are
processed so the initialization only needs to bootstrap the
processing. Experiments have been
run using newstories, the Enron email corpii, SNORT rules, and the Common Attack Pattern
Enumeration and Classification (CAPEC) database.
|
|
Command and Control Systems Engineering Support
|
Navy Program Executive Office for Integrated Warfare Systems
(PEO IWS) 6.0
|
Silver Bullet provided systems engineering support for Navy ship
and submarine navigation systems including inertial, multi-source,
alternative, and decision support.
Silver Bullet developed architecture descriptions and data models,
acquisition documents, and specifications. Silver Bullet conducted mathematical analyses,
supports scientific research, and analyzes and synthesizes operational
requirements. Silver Bullet
provides liaison between engineering and S&T communities. Silver Bullet also supported Combat
System and C4I systems integration and architecture, particularly for
Ballistic Missile Defense (BMD).
|
|
DoD Information Enterprise Architecture
|
Department of Defense (DoD) Chief Information Officer (CIO)
|
Silver Bullet is DoD’s lead engineer for the DoD
Architecture Framework (DoDAF).
Silver Bullet was key in moving DoD architectures from, 1)
unstructured documents to structured data, 2) qualitative to quantitative, and 3) templates to fit-for-purpose
models. This derived from Silver
Bullet's role as the U.S. DoD technical lead to an international defense
team that developed a formal ontology for coalition data exchange called
International Defence Enterprise Architecture Specification (IDEAS). It
is formal in that it has a mathematical foundation that implements high
order type theory and four-dimensional mereotopology. Silver Bullet used
this ontology to develop the DoD Architecture Framework (DoDAF) Meta
Model (DM2), in the Cyber Ontology (CybOnt) project, in the operationalization of the DoD Information Enterprise
Architecture (DIEA), and in the architectures for the Joint Artificial
Intelligence Center (JAIC) Common Foundation.
|
|
Architecture and Systems Engineering
|
Space and Naval Warfare Systems Command
|
Architecture/systems engineering services for the development
of the Navy/Department of Navy Enterprise Architecture (N/DoN EA).
Silver Bullet developed the data structure for the Naval
Architecture Elements Reference Guide (NAERG) using DM2 and supported the
analysis and restructuring of the existing architecture elements into the
DM2. Silver Bullet developed an
OWL version of the DM2.
|
|
Enterprise and Strategic Engineering
|
Joint Improvised Explosive Device – Defeat Organization
(JIEDDO) Counter IED Operations – Intelligence Center (COIC)
|
Silver Bullet is the EA team lead as part of the Systems
Engineering and Technical Assistance (SETA) contractor for the COIC’s
Enterprise Engineering Board.
Silver Bullet develops and maintains the JIEDDO-COIC Enterprise
Architecture “As Is” and “To Be” baselines for capabilities, operations,
systems, services, data, and standards including all COIC system
interfaces, standards, and major technology insertion changes. Silver
Bullet develops and maintains the EA Roadmap from the “As Is” to the “To
Be” state. This includes
supporting mission analysis and identification of gaps to address in “To
Be” EA end state, ensuring alignment with DOD and IC EA (e.g., JIE, DI2E
and ICITE initiatives), supporting COIC portfolio management (PfM).
Silver Bullet also conducts research into advanced analytic techniques
(e.g., network analysis, search, predictive
analysis) and technologies (e.g., electronics, forensics) and the
information technology (e.g., software applications, databases and
advanced database management).
|
|
