Securboration, Inc.

Securboration Awarded Army Contract for Automatic Redacting Capability

Securboration has been awarded a US Army contract for automatic redaction of information based on each requestor’s allowed access level. Existing Army technologies as well as open source alternatives that are interchangeable will be used to provide a computing environment that allows Warfighters access to data that would normally be inaccessible due to the lack of identity and attribute awareness in these systems. Currently, analysts are unable to keep pace and products are incorrectly marked or over classified. The DOD needs the ability to accelerate intelligence dissemination across security enclaves to support the tempo of modern warfare. Dynamic data redaction and filtering of granular intelligence data based on user attributes is required. The objective is to develop a standards-based Identity and Access Management (IdAM) and Attribute-Based Access Control (ABAC) computing framework that is capable of fined-grained access control of metadata queries and product retrievals from the current system.

Securboration Wins Navy Contract for Course of Action Recommender

Securboration has been awarded a contract from the Navy to build a general framework for a course of action (COA) recommender. Securboration will develop COA Recommendation Services (COARS), which is a service-based implementation which fits well with the Navy’s modular approach to maturing leading edge research towards a deployable COA recommender. Securboration is teaming with Dartmouth College and their renowned computational modeling professor, Dr. Eugene Santos, to specifically address predictive capability for COA recommendations. We will extend Dr. Santos’ research in Bayesian Knowledge Bases (BKBs) by adding deep learning concepts to fuse knowledge fragments and derive complex emergent behavior that is missed with existing reasoning techniques and algorithms. This accounts for the intricacies and dynamics of the operational environment and overcomes the limitations of current COA recommender approaches that are based on simplistic assumptions regarding adversary objectives, limitations, and intent.

Securboration Wins Navy Contract for Electronic Warfare Scoring System

Securboration has been awarded a contract from the Navy to develop an automated information extraction and scoring system for Electronic Warfare (EW) systems. Securboration will develop the Maritime Electronic Warfare (MEW) Sustainment Information Card (MEWSIC) system to meet the requirements of this effort. Existing systems require manual research that is hampered by inaccurate information in inventory systems and/or by a partial or limited view of the inter-dependencies upon systems, components, and parts. Much of the necessary information is in textual resources and this effort aims to provide a cost effective and timely method to determine if the proper support solution is available to reduce repair time and maintain its equipment in an optimal operational readiness state. MEWSIC aims to provide this capability by exploiting both structured and unstructured data sources to gain a true picture of a system’s (e.g. ship) operational status and risk factors. The core technology behind MEWSIC is Securboration’s proven integration and text analytics pipeline, which includes distinct capabilities for Entity Extraction, Document Summarization, and Topic Identification and Ranking.

Air Force Awards Contract to Securboration to Enhance Analyst Capabilities

Securboration has been selected by the Air Force Research Laboratory (AFRL) to explore, systematically study and develop technology that enables computers to proactively support analyst activity. The goal is to provide a bidirectional human and information technology partnership where the computer learns, adapts and informs analysts based on their typical practices, situational context, and available data content. The system under development, referred to as the Adaptive Data Immersion Environment or ADIEN, combines machine learning, text analytics, and semantic search to learn and evolve user preferences based on individual usage patterns. Additionally, as new data products become available they will be automatically identified and seamlessly added into analyst’s workspace. ADIEN  is designed to act as a ‘buddy agent’ that not only stays one step ahead of the analyst but does so in an intelligent fashion.  In this capacity, ADIEN provides intelligent offloading to proactively provide assistance to the analyst while they are focused on another task; data immersion to provide the analyst with a wide view of the data landscape allowing them to navigate, test and evaluate hypothesis; continuity in order to maintain the analysis state, allowing analysts to take necessary breaks and seamlessly continue where they left off; and allows for collaboration across analyst networks. ADIEN combines machine learning, text analytics, and semantic search to not only learn and evolve preferences based on individual usage patterns, but also to identify related data relevant to what the analyst is working on and to proactively and seamlessly insert them into the analyst’s workspace. ADIEN’s capabilities include (1) letting the data speak for itself by guiding analysts to where the data convolves, but also highlighting ‘outliers’ of interest to the analyst; (2) continually adapting to and augmenting analyst’s workflow; (3) accommodating the analyst’s interrupt-driven environment and (4) subtly facilitating collaboration without mandating it.

Securboration to Develop Cognitive Knowledge-Aided Interface

Securboration has been awarded a contract by the U.S. Army to develop a cognitive knowledge-aided interface along with supporting information processing techniques to exploit very large data streams and autonomously highlight areas of interest for personnel with limited or no prior knowledge of the area. Securboration will build the Augmented Reality for Tactical Edge Analysis, ARTEA, system to satisfy the requirements. ARTEA will apply deep machine learning and noise reduction algorithms to big-data social media streams to identify patterns, extract attitude trends and shifts, and present the most salient features of the derived knowledge immediately to the edge soldier via non-intrusive augmented reality devices such as Google Glass. ARTEA will use a new approach to knowledge discovery that will flatten the intelligence hierarchy by using social media to enable the edge soldier to act as a vital, active participant in the analysis process.

Securboration develops Collaborative Event Processing Environment (CEPE) for USTRANSCOM

Securboration has been awarded a contract to develop the Collaborative Event Processing Environment (CEPE) for USTRANSCOM. CEPE will wrap current Fusion Center Knowledge Management (KM) capabilities in a graphical user interface (GUI) allowing direct Subject Matter Expert (SME) editing of ontologies and topics, predictive assessment of the meta-tagging and routing impact of these edits, and configuration management of Fusion Center knowledge representation resources. CEPE will be a browser-based tool compatible with the Fusion Center’s existing network and content management footprint, and will capture user feedback and expertise, learning preferences and domain knowledge over time. Rather than attempt to turn SMEs into ontologists, CEPE will bring many repeatable, automatable knowledge representation processes under the SMEs’ control, utilizing a wizard type approach and learning over time to usefully match feedback patterns to new knowledge and improved meta-tagger and message routing behavior.

Securboration Wins Air Force Contract for Sensitive Data Disclosure Risk Calculation System

Securboration has been awarded a contract by the Air Force Research Laboratory (AFRL) to develop algorithms to automatically categorize and quantify the security risks from disclosure of information. It is important to accurately assess the risk associated with different types of information so that protection effort is focused on the most potentially damaging leaks. For some types of information, such as passwords or encryption keys, the consequences of leaking even a few bits of information can be very severe. Securboration will develop the RIsk QUantification and ESTimation (RIQUEST) system to extend the state-of-the art in automatic risk disclosure quantification by applying probabilistic inference to information automatically extracted from open sources.  This effort will focus on three areas that are essential for auto-risk disclosure: Text Analytics, Probabilistic Inference; and Causal Analysis. By making technological advancements in each, Securboration will create a unique and innovative system that performs auto-risk disclosure by ingesting data from multiple heterogeneous data streams, extracting information pertinent to the a piece of sensitive information from those streams, quantifying impacts that the discovered information have on the overall security posture of the sensitive information, and providing complete traceability and transparency into the disclosure risk causes for sensitive information.

Securboration to develop Cross-Organizational Semantic Services (CrOSS) for the Joint Planning and Development Office (JPDO)

Securboration, working with the JPDO and AFRL, has been funded to extend the Ontology Generation and Evolution Processor (OGEP)research and development effort into an operational CrOSS capability.  CrOSS will expedite the automatic processing of both formal planning, programming, and acquisition information sources (for example, enterprise architecture (EA), requirements, research and development (R&D) roadmaps, etc…) and informal artifacts (textual documents, reports, statistics, …) to help identify and synchronize interagency activities. This effort will help drive interagency synchronization to more efficiently deliver NextGen capabilities as the DoD seeks “order of magnitude” NextGen cost reductions for the Department of Defense (DoD), Federal Aviation Administration (FAA), and partners.

Securboration’s Requirements Analysis Technology Deployed at United States Transportation Command

Securboration, Inc. delivered the latest version of their Requirements Analysis Portlet (RAP) to the United States Transportation Command (USTRANSCOM) Common Production Environment (CPE). RAP addresses the organization’s need to map a requirement to joint capability areas, capability gaps, and organizational activities based on unstructured text in a needs statement.  RAP uses Securboration’s Semantic Grounding Mechanism (SGM) to provide a common understanding of a system’s capabilities and data requirements that is based on semantics instead of rigid syntax.  It provides accurate mapping suggestions in a timely fashion, freeing the operator from time consuming manual mapping processes.

Securboration develops Collaborative Event Processing Environment (CEPE) for USTRANSCOM

Securboration has been awarded a contract to develop the Collaborative Event Processing Environment (CEPE) for USTRANSCOM. CEPE will wrap current Fusion Center Knowledge Management (KM) capabilities in a graphical user interface (GUI) allowing direct Subject Matter Expert (SME) editing of ontologies and topics, predictive assessment of the meta-tagging and routing impact of these edits, and configuration management of Fusion Center knowledge representation resources.  CEPE will be a browser-based tool compatible with the Fusion Center’s existing network and content management footprint, and will capture user feedback and expertise, learning preferences and domain knowledge over time.  Rather than attempt to turn SMEs into ontologists, CEPE will bring many repeatable, automatable knowledge representation processes under the SMEs’ control, using wizarding and learning over time to usefully match feedback patterns to new knowledge and improved meta-tagger and message routing behavior.

« Previous PageNext Page »

Securboration, Inc.