Results for “Analyst” — Page 22

NASA NTRS2017

Collaborative, Rapid Mapping of Water Extents During Hurricane Harvey Using Optical and Radar Satellite Sensors

On August 25, 2017, Hurricane Harvey made landfall between Port Aransas and Port O'Connor, Texas, bringing with it unprecedented amounts of rainfall and record flooding. In times of natural disasters of this nature, emergency responders require timely and accurate information about the hazard in order to assess and plan for disaster response. Due to the extreme flooding impacts associated with Hurricane Harvey, delineations of water extent were crucial to inform resource deployment. Through the USGS's Hazards Data Distribution System, government and commercial vendors were able to acquire and distribute various satellite imagery to analysts to create value-added products that can be used by these emergency responders. Rapid-response water extent maps were created through a collaborative multi-organization and multi-sensor approach. One team of researchers created Synthetic Aperture Radar (SAR) water extent maps using modified Copernicus Sentinel data (2017), processed by ESA. This group used backscatter images, pre-processed by the Alaska Satellite Facility's Hybrid Pluggable Processing Pipeline (HyP3), to identify and apply a threshold to identify water in the image. Quality control was conducted by manually examining the image and correcting for potential errors. Another group of researchers and graduate student volunteers derived water masks from high resolution DigitalGlobe and SPOT images. Through a system of standardized image processing, quality control measures, and communication channels the team provided timely and fairly accurate water extent maps to support a larger NASA Disasters Program response. The optical imagery was processed through a combination of various band thresholds and by using Normalized Difference Water Index (NDWI), Modified Normalized Water Index (MNDWI), Normalized Difference Vegetation Index (NDVI), and cloud masking. Several aspects of the pre-processing and image access were run on internal servers to expedite the provision of images to analysts who could focus on manipulating thresholds and quality control checks for maximum accuracy within the time constraints. The combined results of the radar- and optical-derived value-added products through the coordination of multiple organizations provided timely information for emergency response and recovery efforts.

SERVIR↗

NASA NTRS2017

Program to Optimize Simulated Trajectories II (POST2) Surrogate Models for Mars Ascent Vehicle (MAV) Performance Assessment

The primary purpose of the multiPOST tool is to enable the execution of much larger sets of vehicle cases to allow for broader trade space exploration. However, this exploration is not achieved solely with the increased case throughput. The multiPOST tool is applied to carry out a Design of Experiments (DOE), which is a set of cases that have been structured to capture a maximum amount of information about the design space with minimal computational effort. The results of the DOE are then used to fit a surrogate model, ultimately enabling parametric design space exploration. The approach used for the MAV study includes both DOE and surrogate modeling. First, the primary design considerations for the vehicle were used to develop the variables and ranges for the multiPOST DOE. The final set of DOE variables were carefully selected in order to capture the desired vehicle trades and take into account any special considerations for surrogate modeling. Next, the DOE sets were executed through multiPOST. Following successful completion of the DOE cases, a manual verification trial was performed. The trial involved randomly selecting cases from the DOE set and running them by hand. The results from the human analyst's run and multiPOST were then compared to ensure that the automated runs were being executed properly. Completion of the verification trials was then followed by surrogate model fitting. After fits to the multiPOST data were successfully created, the surrogate models were used as a stand-in for POST2 to carry out the desired MAV trades. Using the surrogate models in lieu of POST2 allowed for visualization of vehicle sensitivities to the input variables as well as rapid evaluation of vehicle performance. Although the models introduce some error into the output of the trade study, they were very effective at identifying areas of interest within the trade space for further refinement by human analysts. The next section will cover all of the ground rules and assumptions associated with DOE setup and multiPOST execution. Section 3.1 gives the final DOE variables and ranges, while section 3.2 addresses the POST2 specific assumptions. The results of the verification trials are given in section 4. Section 5 gives the surrogate model fitting results, including the goodness-of-fit metrics for each fit. Finally, the MAV specific results are discussed in section 6.

Zwack, M. R.↗

NASA NTRS2019

Space Launch System and Missions to Near Rectilinear Halo Orbits

The first major evolution of NASA’s Space Launch System (SLS) will begin its flights starting in the mid-2020’s. This new configuration, called Block 1B, replaces the Interim Cryogenic Propulsion Stage with a larger Exploration Upper Stage (EUS). The additional capability provided by the new upper stage will allow SLS to send heavier payloads into deep space. One destination of interest to the SLS program is called a Near Rectilinear Halo Orbit (NRHO). This is a type of lunar orbit with multiple advantages for deep space exploration. These benefits include Earth/lunar access, low station-keeping requirements, and high communication potential with Earth. Therefore, it is a leading candidate for the proposed Lunar Orbital Platform-Gateway (LOP-G). This paper will provide a detailed assessment of the SLS Block 1B requirements and capabilities for sending payloads to an NRHO. Analysts at Marshall Space Flight Center are producing a multi-year mission availability scan for the SLS Block 1B configuration to a predefined NRHO orbit. The analysis produces an optimized trajectory for each day of the scan window. A maximized payload and minimized propellant requirement are determined for each day. All maneuvers from launch to the end of the Trans-Lunar-Injection (TLI) are being modeled as finite burns. Injections into an NRHO are being modeled as impulsive maneuvers. The payload element is arbitrary, but includes sufficient mass to represent a large habitat or propulsion module. The resulting parameters of payload capability, delta-v requirements, and launch windows length vary over the course of the scan. Many launch days in the scan are eliminated in post-processing as they violate mission constraints such as payload mass and propellant usage to insert into an NRHO. Based off previous one-year scan results for the SLS Block 1B Design Analysis Cycle 2 (DAC-2), it is expected that there will be one to three days per week where the payload is able to insert into an NRHO within SLS constraints objectives. This scan provides results for longer than one-year, allowing analysts to better understand the launch availability and energy requirements of SLS Block 1B over time. The in-space mission design and scans utilize Copernicus, an n-body trajectory optimization tool originally developed out of the University of Texas at Austin with further development at Johnson Space Center in Houston, TX. To seed the in-space trajectory, Copernicus uses a plugin to call a database of SLS ascent trajectories optimized in the Program to Simulate Optimized Trajectories II (POST2). The ascent trajectories are developed using a framework that parametrizes payload mass and LEO inclination, and targets a 100 nmi (nautical mile) altitude circular parking orbit. Though this analysis is specific to the SLS program, it will provide a summary of mission design benefits and constraints associated with generic NRHO access, and may be applied to other programs or concepts that will utilize this orbit.

Statham, T. L.↗

NASA NTRS2018

Generating Exploration Mission-3 Trajectories to a 9:2 NRHO using Machine Learning

The purpose of this thesis is to design a machine learning algorithm platform that provides expanded knowledge of mission availability through a launch season by improving trajectory resolution and introducing launch mission forecasting. The specific scenario addressed in this paper is one in which data is provided for four deterministic translational maneuvers through a mission to a Near Rectilinear Halo Orbit (NRHO) with a 9:2 synodic frequency. Current launch availability knowledge under NASA's Orion Orbit Performance Team is established by altering optimization variables associated to given reference launch epochs. This current method can bean abstract task and relies on an orbit analyst to structure a mission based on an established mission design methodology associated to the performance of Orion and NASA's Space Launch System. Introducing a machine learning algorithm trained to construct mission scenarios within the feasible range of known trajectories reduces the required interaction of the orbit analyst by removing the needed step of optimizing the orbit to fit an expected translational response required of the spacecraft. In this study, k-Nearest Neighbor and Bayesian Linear Regression successfully predicted classical orbital elements for the launch windows observed. However both algorithms had limitations due to their approaches to model fitting. Training machine learning algorithms off of classical orbital elements introduced a repetitive approach to reconstructing mission segments for different arrival opportunities through the launch window and can prove to be a viable method of launch window scan generation for future missions.

Guzman, Esteban↗

NASA NTRSDate not supplied

Methods for Modeling Particle Impingement in Solid Rocket Motor Nozzles

This paper describes several approaches available to analysts to capture the mechanical erosion effects of alumina droplet impingement in solid rocket motor nozzles. Supersonic alumina impingement is of great interest to solid rocket motor analysts, as it can be one of the main contributors to overall nozzle erosion, particularly in nozzles with highly contoured exit cones. An overview of the basic concepts of mechanical impingement is provided, along with discussion on some of the various methods available, ranging from the purely empirical to more physics-based techniques, and their respective advantages and disadvantages.

William Allen Till↗

NASA NTRSDate not supplied

NESC GN&C TDT Workshop on 2D Image Motion Optical Transfer Functions, Pointing Performance Analysis, and Requirements

This course was prepared by the author in support of the NASA GN&C Technical Discipline Team (TDT). The purpose of this two-day course is to inform practicing GN&C engineers, and other system and subsystem engineers, including payload engineers and mission analysts, of methodologies to analyze the pointing performance of spacecraft and to write unambiguous pointing requirements that are relevant to optical payload performance. It is hoped that the seminar will contribute to a best-practices manual and contribute to accepted and uniform means of requirements definition, validation, and verification. The course material is designed to address the needs of not only practicing GN&C engineers, but also system and subsystem engineers of other disciplines, payload engineers, mission analysts, and even astronomers who need to understand how they interface to a GN&C subsystem. Much revision is the direct result of active student participation during the presentations and from feedback obtained through course evaluation forms.

2D Image↗

NASA NTRSDate not supplied

A Survey Protocol to Assess Meaningfulness and Usefulness of Automated Topic Finding in the NASA Aviation Safety Reporting System

Context: The NASA Aviation Safety Reporting System (ASRS) is a voluntary confidential aviation safety reporting system. The ASRS receives reports from pilots, air traffic controllers, flight attendants and other involved in aviation operations. The reports are de-identified and coded by ASRS expert safety analysts and a short descriptive synopsis is written to describe the safety issue. The de-identified reports are then disseminated to the aviation community in a number of ways including entry into an online database, Safety Alert Bulletins and For Your Information Notices, and the CALLBACK newsletter. Key to these publications are the timely processing (de-identification, coding and summarization) of new reports, which is currently done by ASRS expert safety analysts. Thus, we believe topic modelling could decrease effort in ASRS, if topics are comprehensible. Aim: We propose a methodology to evaluate whether automated topic finding using topic modelling provides meaningful and useful topics. Method: We extend the total error survey methodology to evaluate user topic comprehension of machine learning outputs. To accomplish this we performed a literature review to identify existing methods and define a construct for topic comprehension, utilizing existing ASRS synopsis writing practices to more precisely define meaningfulness and usefulness. Results: A survey protocol was created that addresses the limitations of other survey protocols found in the literature review, which we found lacking in rationale and clear protocol definition. Conclusion: The surveying of user understanding in machine learning outputs presents challenges due to the explosion of parameters to control for and the lack of systematic approach presented in the literature. More reproducible work and survey protocols are needed in the literature and our work is one step towards that direction.

topic finding↗

NASA NTRSDate not supplied

Exploitation of a Validation Hierarchy for Modeling and Simulation

Across engineering there is an evolving need to increase reliance on physics-based simulation to develop, design and optimize engineering systems. This increased reliance on modeling and simulation has highlighted a growing need to transform the confidence that modeling and simulation analysts have in their results into credibility for systems engineers to design and field systems more quickly and with less physical testing. For isolated components of a complex system, where a single discipline may drive product design, this is less of a concern as the relationship is often straightforward and easy to explain. However, when these isolated components are integrated, and are expected to operate in a multi-disciplinary context in which safety critical systems are involved, new concepts and model assurance standards are required. In this paper we address this challenge by showing how a model validation hierarchy can be exploited to identify those model validation experiments that will contribute most to increasing confidence and credibility of modeling and simulation predictions. The approach that is adopted contains four main steps. The first step is the construction of a model validation hierarchy that links subsystems, assemblies, and components to a hierarchy of physical experiments that can be used support model validation. This hierarchy connects the concerns of systems engineers to those of the modeling and simulation analyst in a clear and logical way. The structure and content of this hierarchy is then used in a second step to establish which physical phenomena have the greatest impact on overall system performance metrics. A gap analysis technique, based upon modeling and simulation concerns, is then used to prioritize the important physical phenomenon. Unfortunately, a common outcome of such gap analyses is the identification of many important gaps and so, in the final step of our process, we advocate the use of a global sensitivity analysis as a means to complete the prioritization.

Verification and Validation↗

NASA NTRSDate not supplied

Exploitation of a Validation Hierarchy for Modeling and Simulation

Across engineering there is an evolving need to increase reliance on physics-based simulation to develop, design and optimize engineering systems. This increased reliance on modeling and simulation has highlighted a growing need to transform the confidence that modeling and simulation analysts have in their results into credibility for systems engineers to design and field systems more quickly and with less physical testing. For isolated components of a complex system, where a single discipline may drive product design, this is less of a concern as the relationship is often straightforward and easy to explain. However, when these isolated components are integrated, and are expected to operate in a multi-disciplinary context in which safety critical systems are involved, new concepts and model assurance standards are required. In this paper we address this challenge by showing how a model validation hierarchy can be exploited to identify those model validation experiments that will contribute most to increasing confidence and credibility of modeling and simulation predictions. The approach that is adopted contains four main steps. The first step is the construction of a model validation hierarchy that links subsystems, assemblies, and components to a hierarchy of physical experiments that can be used support model validation. This hierarchy connects the concerns of systems engineers to those of the modeling and simulation analyst in a clear and logical way. The structure and content of this hierarchy is then used in a second step to establish which physical phenomena have the greatest impact on overall system performance metrics. A gap analysis technique, based upon modeling and simulation concerns, is then used to prioritize the important physical phenomenon. Unfortunately, a common outcome of such gap analyses is the identification of many important gaps and so, in the final step of our process, we advocate the use of a global sensitivity analysis as a means to complete the prioritization.

Verification and Validation↗

NASA NTRS2023

CST: A Tool for Optimizing the Efficiency and Effectiveness of Static-Code Analysis Tools

Static Code Analysis (SCA) is a vital component of NASA IV&V’s mission assurance for safety-critical software as it reduces the likelihood of software-induced hazards impacting mission success. Static Code Analysis achieves this by identifying hazards that may not have been otherwise detectable by typical code reviews or other testing. Using SCA tools, however, can be intimidating due to steep learning curves, especially considering tool performance and defect coverage varies greatly. Because of this variation amongst SCA tools, understanding which tools support certain defects and which do not, as well as understanding how to run an analysis based on steps that are unique to each tool, can be difficult to both new and experienced analysts alike. To mitigate this, the SCAWG or the IV&V Static Code Analysis Working Group, created the SCA Checker Taxonomy and Starting Point Profiles. The Checker Selection Tool (CST) incorporates these two SCAWG products into an interactive tool which allows the user to: select organized categories of defects they would like the SCA tools to discover, select default checkers depending on their mission type (e.g. flight), and configure multiple SCA tools at once. C/C++, Java, and Python defect checkers from four common SCA tools were utilized in this iteration of the CST. This iteration also includes the addition of training, SCA tool specific help, and taxonomy guide links, into its design to help users new to Static Code Analysis learn how to perform SCA more efficiently. The CST has been subject to beta testing by experienced static code analysts from the SCAWG to ensure a usable and accurate final product. The implications of the CST in the mission assurance of NASA safety-critical software are profound, as the CST can help identify and reduce false positives and false negatives, fundamentally improving overall SCA efficiency and accuracy.

static code analysis↗

NASA NTRSDate not supplied

Safety Expertise and the Perils of Novelty

Emerging aviation markets such as urban air mobility are giving rise to new technologies and means of operation. However, novelty may hide ‘unknown unknowns,’ raising new hazards. This paper examines how expertise and safety techniques enable transformative technologies such as reduced crew operations, hybrid wing-borne and rotor-born flight, federated air traffic services, and urban operations. We explore how analysts use expertise to address common-cause failures, collect and interpret safety data, and perform exacting tradeoffs between dissimilarity, redundancy, independence, and diversity (human, process lifecycle, or otherwise) to ensure safety. When novelty is present, analysts might not possess the expertise needed to fully understand the implications of design decisions and tradeoffs being made, especially in early lifecycle phases, on emergent properties such as safety. Safety expertise must be carefully cultivated. The conflicting views of safety experts must be unpacked to identify the divergence in fundamental assumptions, models, means, and methods that may be causing them. Once systems venture beyond the basis of what safety expertise can reliably guarantee, projects take on risk that must be managed. The paper contains key takeaways and actionable recommendations for novel OEMs and regulators touching on topics such as robust monitoring; clear and transparent reporting; incremental approaches to fielding novel systems in hazard-rich, risk-tolerant environments; the cultivation of safety culture and expertise in an organization; and the use of scientific study to reduce epistemic uncertainty in novel operations with new technologies. Since excessive novelty in aviation can undermine the current foundation of safety, humility and incrementalism are necessary to enable emerging aviation markets safely.

safety expertise↗

NASA NTRSDate not supplied

Safety Expertise and the Perils of Novelty

Emerging aviation markets such as urban air mobility are giving rise to new technologies and means of operation. However, novelty may hide ‘unknown unknowns,’ raising new hazards. This paper examines how expertise and safety techniques enable transformative technologies such as reduced crew operations, hybrid wing-borne and rotor-born flight, federated air traffic services, and urban operations. We explore how analysts use expertise to address common-cause failures, collect and interpret safety data, and perform exacting tradeoffs between dissimilarity, redundancy, independence, and diversity (human, process lifecycle, or otherwise) to ensure safety. When novelty is present, analysts might not possess the expertise needed to fully understand the implications of design decisions and tradeoffs being made, especially in early lifecycle phases, on emergent properties such as safety. Safety expertise must be carefully cultivated. The conflicting views of safety experts must be unpacked to identify the divergence in fundamental assumptions, models, means, and methods that may be causing them. Once systems venture beyond the basis of what safety expertise can reliably guarantee, projects take on risk that must be managed. The paper contains key takeaways and actionable recommendations for novel OEMs and regulators touching on topics such as robust monitoring; clear and transparent reporting; incremental approaches to fielding novel systems in hazard-rich, risk-tolerant environments; the cultivation of safety culture and expertise in an organization; and the use of scientific study to reduce epistemic uncertainty in novel operations with new technologies. Since excessive novelty in aviation can undermine the current foundation of safety, humility and incrementalism are necessary to enable emerging aviation markets safely.

safety expertise↗

NASA NTRSDate not supplied

Deep Neural Network Based Convergence Classification for Computational Fluid Dynamics

A supervised deep learning approach is coupled with heuristic convergence criteria to construct a classification model for detecting the completion (convergence) of computational fluid dynamics (CFD) simulations. Heuristic convergence criteria alone are not always sufficient and more complex decisions are often left to a human analyst. The proposed approach leverages heuristic convergence criteria as well as two deep neural network (DNN) models, one binary and one multi-class, to improve the efficiency and consistency of convergence classification across a wide range of flight regimes. The DNN models presented are each trained on a subset of ascent aerodynamic CFD simulations for NASA’s Space Launch System and were produced using NASA’s unstructured Navier-Stokes solver FUN3D. Individual solutions are analyzed intermittently and are classified as sufficiently converged, further iterations required, or switch from steady Reynolds Averaged Navier-Stokes (RANS) to unsteady RANS CFD based on the iterative histories of four aerodynamic coefficients. The implemented classification model is shown to produce solutions that closely correlate to solutions produced by a human analyst. This work lays groundwork for expanding the capabilities of DNNs for automating and improving more of the CFD process.

SLS↗

NASA NTRS2012

Structural Analyses and Margins of Safety

There is an increasing reliance on modeling and simulation to verify, quantify, and certify designs of complex structures. The availability of a range of commercial modeling and simulations tools and packages with a variety of capabilities, in conjunction with increased computational resources, is allowing analysts to rapidly perform detailed analyses. However, care should be taken to understand specific tool limitations, assumptions, and boundary conditions as erroneous results can be generated without being recognized by the analysts. In addition, the reported margins of safety should be carefully interrogated to identify any false positive or negative margins and highlight any areas for structural concern.

Structural Analyses↗

NASA NTRSDate not supplied

Global Sensitivity Analyses for Test Planning with Black-Box Models for Mars Sample Return

This work describes sensitivity analyses performed on complex black-box models used to support experimental test planning under limited resources in the context of the Mars Sample Return program, which aims at bringing to Earth rock and atmospheric samples from Mars. We develop a systematic workflow that allows the analysts to simultaneously obtain quantitative insights on key drivers of uncertainty, on the direction of impact, and the presence of interactions. We apply novel optimal transport-based global sensitivity measures to tackle the multivariate nature of the output. On the modeling side, we apply multi-fidelity techniques that leverage low-fidelity models to speed up the calculations and make up for the limited amount of high-fidelity samples, while keeping these in the loop for accuracy guarantees. The sensitivity analysis reveals insights useful for the analysts to understand the model's behavior and identify the factors to focus on during testing in order to maximize the value of information extracted from them to ensure mission success when limited resources are available.

Giuseppe Cataldo↗

NASA NTRS2024

Automated Detection of Spurious Signals in VLBI Phase Calibration Data

In this memorandum, a set of processing strategies for automatic masking of phase calibration tones is outlined as implemented in the software package PIMA in the task Generate Phase Calibration Mask, or GEPM. The task relies on a robust procedure of cleaning phase calibration data before employing several mathematical strategies designed to selectively identify spurious signals from phase calibration tones. These strategies were derived as more rigorous implementations of heuristics traditionally used by analysts in manually identifying problematic phase calibration data. The task is intended to automate the process of generating a phase calibration mask and in so doing increase the speed and regularity of VLBI analysis. At the outset of the project, a series of goals were identified to evaluate the success of this mask generation. This included the development of an algorithm to identify and mask short-term (defined as less than 10 seconds in length) spurious signals affecting phase calibration data, the development of an algorithm to identify and mask phase calibration tones affected by constant radio-frequency interference, the implementation of a detection scheme for identifying large jumps in phase calibration phase caused by clock breaks, and finally a method of reporting phase calibration health metrics to the user. Processing of over 10 experiments has demonstrated that the final form of the task GEPM as detailed in this technical memorandum satisfies each of these conditions and satisfactorily performs the task of automatic phase calibration data masking, although the wide variety in quality and characteristics of phase calibration data makes a single solution to the problem quite difficult. A series of user inputs have therefore been defined to assist analysts in tailoring automatic masking to specific stations and phase calibration generators. The vast majority of applicable code was written in FORTRAN to increase execution speed and ease of interfacing with the existing code base in PIMA, but a wrapper function in Python was also written to allow for a simpler method of interacting with GEPM and inputting relevant parameters. In addition to this report, thorough documentation was added to the already existing repositories associated with PIMA as a whole.

Geodesy↗

NASA NTRS2026

Development, Characterization, and Validation of Elevated-Temperature Constitutive Models: Deformation and Damage

This report provides a brief review of experimentally observed hereditary and nonhereditary material behavior along with background information on standard as well as advanced internal state variable constitutive modeling at elevated temperature. A description of exploratory, characterization, and validation testing is presented along with a detailed outline of what constitutes “sufficient” data content (i.e., quality and quantity) for developing or enhancing, characterizing, and validating a particular sophisticated nonlinear time- and history-dependent (hereditary) class of constitutive models known as GVIPS (generalized viscoplasticity with potential structure). The tests described are necessary to reveal a material’s behavior in the reversible (or viscoelastic) and irreversible (or viscoplastic) regimes, for the identification of both deformation and damage model parameters. Results presented are primarily for metallic materials. In all cases, both uniaxial and multiaxial tests are described, and the linkage between the specific tests and the parameters within the model that can be characterized from the results of these tests are also discussed. Discussion is also provided relative to the role information management must play relative to material data collection, analysis, maintenance, and dissemination. The need for such an information system is particularly important as both analyst and designer move toward utilizations of sophisticated, nonlinear time- and history dependent (hereditary) constitutive models. Lastly, the concept of state space and its utility in understanding and establishing constitutive models is addressed throughout. The intent behind this document is to help both the modeler and experimentalist understand each other’s specific points of view and provide guidance for both model development and characterization. Emphasis has been placed on providing the mechanician with information regarding how tests are performed and what issues to be aware of when interpreting results. It is hoped that experimentalists will take away a new perspective on the types of information that modelers and analysts are looking for from them.

Constitutive Modeling↗

DOE OSTI2012

Guided Text Analysis Using Adaptive Visual Analytics

This paper demonstrates the promise of augmenting interactive visualizations with semi-supervised machine learning techniques to improve the discovery of significant associations and insight in the search and analysis of textual information. More specifically, we have developed a system called Gryffin that hosts a unique collection of techniques that facilitate individualized investigative search pertaining to an ever-changing set of analytical questions over an indexed collection of open-source publications related to national infrastructure. The Gryffin client hosts dynamic displays of the search results via focus+context record listings, temporal timelines, term- frequency views, and multiple coordinated views. Furthermore, as the analyst interacts with the display, the interactions are recorded and used to label the search records. These labeled records are then used to drive semi-supervised machine learning algorithms that re-rank the unlabeled search records such that potentially relevant records are moved to the top of the record listing. Gryffin is described in the context of the daily tasks encountered at the Department of Homeland Securitys Fusion Centers, with whom we are collaborating in its development. The resulting system is capable of addressing the analysts information overload that can be directly attributed to the deluge of information that must be addressed in search and investigative analysis of textual information.

machine learning↗