Search NASA⌕ Search

SEARCH · Search NASA

Results for “subset sampling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 181 records · Page 10

The kinematics of dense clusters of galaxies. I - The data

We have measured redshifts in the fields of the 31 R greater than 1 Abell clusters with z of 0.02-0.05 and absolute b exp II above 30 deg. At least ten of the fields are severely contaminated by superimposed velocity peaks. We derive the mean velocities and velocity dispersions of the 25 dense peaks in the sample. The abundance of peaks, 6.6 x 10 exp -6 h-cubed Mpc exp -3, is consistent with the mean number density of R above 1 Abell clusters. The range of velocity dispersions is 304-1346 km/s. The median dispersion is 718 km/s. The subset of eight systems with cD galaxies has a median velocity dispersion of 792 km/s, close to that of non-cD systems (626 km/s). When these data are combined with 16 cD cluster velocity dispersions from our previous study (Zabludoff et al., 1990, or ZHG) and Dunn (1991), eight of 25 cD galaxies have peculiar motions larger than half the cluster velocity dispersions. These findings further support the conclusions of Beers and Geller (1983), ZHG, and Dunn (1991), who argue that cD galaxies do not lie in the global kinematic center, but in local potential minima. If so, systems with speeding cD's are probably a guide to substructure in dynamically evolving systems.

Zabludoff, Ann I.↗

Time-dependent analysis of 8 days of CN spatial profiles in comet P/Halley

CN profiles in comet P/Halley were constructed from observations taken at three observatories during an 8 day period in April 1986. These data provide a time series of CN spatial profiles spanning just over one 7.37 day period from 1986 April 7 to April 15 and sample distances from the nucleus from just over 10(exp 3) km to 10(exp 6) km. The effect of the 7.37 day periodic variation on the CN distribution in P/Halley has been examined by using the time-dependent model applied earlier to a subset of the data. Because of the large spatial scale of the data on April 7, 8, and 9 (approx. 10(exp 6) km), and the corresponding transport time in the coma, information present in the spatial profiles regarding the gas production rate actually covers nearly two full periods. These spatially extended profiles clearly show the wavy structures outside 10(exp 5) km. Such structures were predicted in a previous analysis (Combi & Fink 1993) that was based solely on the photometric light curve and on profiles which only extended to distances less than 10(exp 5) km. We are now able to reproduce the highly variable Halley correction for the variation in gas production rate.

Combi, Michael↗

Probabilistic Assessment of Tunguska-Scale Asteroid Impacts

The Tunguska meteor airburst that felled trees across >2000 sq km of Siberian forest in 1908 has been extensively studied and modeled in attempts to deduce its size, properties, and impact characteristics. However, most of the existing modeling and simulation studies have investigated a small subset of cases based on assumptions of representative densities, velocities, or other properties. In this study, we use the Probabilistic Asteroid Impact Risk (PAIR) model to assess 50 million Tunguska-scale asteroid impacts, covering a full range of potential impactor properties. The impact cases are sampled from probabilistic distributions representing our current knowledge of asteroid properties, entry trajectories, and size frequencies, and the entry, airburst, and resulting ground damage are modeled for each case. The results provide a broader characterization of the range and relative likelihood of asteroid properties that could yield Tunguska-scale impacts. The full results of this study and a companion study on impact frequencies are pending publication in an upcoming Tunguska special edition of Icarus [1,2].

Wheeler, Lorien↗

The numerical evaluation of the maximum-likelihood estimate of a subset of mixture proportions

Necessary and sufficient conditions are given for a maximum likelihood estimate of a subset of mixture proportions. From these conditions, likelihood equations are derived satisfied by the maximum-likelihood estimate and a successive-approximations procedure is discussed as suggested by equations for numerically evaluating the maximum-likelihood estimate. It is shown that, with probability one for large samples, this procedure converges locally to the maximum-likelihood estimate whenever a certain step-size lies between zero and two. Furthermore, optimal rates of local convergence are obtained for a step-size which is bounded below by a number between one and two.

Peters, B. C., Jr.↗

Landsat-based estimation of California's irrigated land

The procedure developed uses two-phase sampling, stratification and multidate Landsat imagery to produce the estimate. Maximizing the advantages of both spectral data and field patterns available from Landsat, three dates of Landsat imagery are interpreted to provide county-wide estimates of the proportion of land irrigated. Ground data, collected on a subset of the area interpreted on Landsat, are used to calibrate the satellite estimate. The satellite and ground measurements are reduced to proportion data and linked using a regression estimator to produce the estimate of irrigated land. It is estimated that 3.99 million hectares were irrigated with a relative standard error of + or - 1.18% at the 99% confidence level. Acreage tabulations provided by the Department of Water Resources show that the Landsat-based estimate differed by less than 0.4% at the state level.

Wall, S. L.↗

Methods and Results for a Global Precipitation Measurement (GPM) Validation Network Prototype

As one component of a ground validation system to meet requirements for the upcoming Global Precipitation Measurement (GPM) mission, a quasi-operational prototype a system to compare satellite- and ground-based radar measurements has been developed. This prototype, the GPM Validation Network (VN), acquires data from the Precipitation Radar (PR) on the Tropical Rainfall Measuring Mission (TRMM) satellite and from ground radar (GR) networks in the continental U.S. and participating international sites. PR data serve as a surrogate for similar observations from the Dual-frequency Precipitation Radar (DPR) to be present on GPM. Primary goals of the VN prototype are to understand and characterize the variability and bias of precipitation retrievals between the PR and GR in various precipitation regimes at large scales, and to improve precipitation retrieval algorithms for the GPM instruments. The current VN capabilities concentrate on comparisons of the base reflectivity observations between the PR and GR, and include support for rain rate comparisons. The VN algorithm resamples PR and GR reflectivity and other 2-D and 3-D data fields to irregular common volumes defined by the geometric intersection of the instrument observations, and performs statistical comparisons of PR and GR reflectivity and estimated rain rates. Algorithmic biases and uncertainties introduced by traditional data analysis techniques are minimized by not performing interpolation or extrapolation of data to a fixed grid. The core VN dataset consists of WSR-88D GR data and matching PR orbit subset data covering 21 sites in the southeastern U. S., from August, 2006 to the present. On average, about 3.5 overpass events per month for these WSR-88D sites meet VN criteria for significant precipitation, and have matching PR and GR data available. This large statistical sample has allowed the relative calibration accuracy and stability of the individual ground radars, and the quality of the PR reflectivity attenuation correction in convective and stratiform precipitation to be evaluated. We will present results of PR-GR reflectivity and rain rate bias comparisons for each OR site, and for different rain types, for the full data set and as time series. The capabilities of the statistical analysis and vertical cross section tools for display and analysis of individual site overpass event data will be described, and examples of the tools' outputs will be shown.

Morris, Kenneth R.↗

Fermi-LAT Gamma-ray Bursts and Insight from Swift

A new revolution in GRB observation and theory has begun over the last 3 years since the launch of the Fermi gamma-ray space telescope. The new window into high energy gamma-rays opened by the Fermi-LAT is providing insight into prompt emission mechanisms and possibly also afterglow physics. The LAT detected GRBs appear to be a new unique subset of extremely energetic and bright bursts. In this talk I will discuss the context and recent discoveries from these LAT GRBs and the large database of broadband observations collected by Swift over the last 7 years and how through comparisons between the Swift, GBM, and LAT GRB samples, we can learn about the unique characteristics and relationships between each population.

Racusin, Judith L.↗

ImageLabler: Labeling and Managing Image Data for Machine Learning in the Earth Sciences

While machine learning techniques for image classification have been around for a long time, storing and managing the vast number of images required as training data is still a problem for scientists. This is especially true for the field of Earth science, where only recently have experts begun using machine learning techniques for image-based phenomena classification. Image Labeler, a fast and scalable cloud-based tagging platform for Earth science images, seeks to improve upon existing methods of managing images and associated metadata, such as maintaining categorized folders of images on a local machine, a process that can be cumbersome and difficult to scale. The platform facilitates rapid development of image-based Earth science phenomena training datasets by allowing scientists to upload their existing imagery as well as extract new samples from open satellite imagery services made available through NASA’s Global Imagery Browse Service (GIBS). Image Labeler also supports GeoTIFF data, with capabilities such as displaying GeoTIFFs on an interactive map, drawing shapefiles over them, and tagging them with additional metadata. This allows scientists to perform spatiotemporal subsetting with geographic information and develop training data more quickly. Built using modern web technologies, Image Labeler includes additional capabilities such as team collaboration for large-scale image tagging projects. Users can download their data in a machine-learning-ready format, allowing scientists to spend time on experimentation rather than on the collection of training data. In this presentation, we demonstrate how Image Labeler seeks to become a one-stop image data management solution for machine learning applications in Earth science.

Ashish Acharya↗

The HYSPIRI Decadal Survey Mission: Update on the Mission Concept and Science Objectives for Global Imaging Spectroscopy and Multi-Spectral Thermal Measurements

The NASA HyspIRI mission is planned to provide global solar reflected energy spectroscopic measurement of the terrestrial and shallow water regions of the Earth every 19 days will all measurements downlinked. In addition, HyspIRI will provide multi-spectral thermal measurements with a single band in the 4 micron region and seven bands in the 8 to 12 micron region with 5 day day/night coverage. A direct broadcast capability for measurement subsets is also planned. This HyspIRI mission is one of those designated in the 2007 National Research Council (NRC) Decadal Survey: Earth Science and Applications from Space. In the Decadal Survey, HyspIRI was recognized as relevant to a range of Earth science and science applications, including climate: "A hyperspectral sensor (e.g., FLORA) combined with a multispectral thermal sensor (e.g., SAVII) in low Earth orbit (LEO) is part of an integrated mission concept [described in Parts I and II] that is relevant to several panels, especially the climate variability panel." The HyspIRI science study group was formed in 2008 to evaluate and refine the mission concept. This group has developed a series of HyspIRI science objectives: (1) Climate: Ecosystem biochemistry, condition & feedback; spectral albedo; carbon/dust on snow/ice; biomass burning; evapotranspiration (2) Ecosystems: Global plant functional types, physiological condition, and biochemistry including agricultural lands (3) Fires: Fuel status, fire frequency, severity, emissions, and patterns of recovery globally (4) Coral reef and coastal habitats: Global composition and status (5) Volcanoes: Eruptions, emissions, regional and global impact (6) Geology and resources: Global distributions of surface mineral resources and improved understanding of geology and related hazards These objectives are achieved with the following measurement capabilities. The HyspIRI imaging spectrometer provides: full spectral coverage from 380 to 2500 at 10 nm sampling; 60 m spatial sampling with a 150 km swath; and fully downlinked coverage of the Earth's terrestrial and shallow water regions every 19 days to provide seasonal cloud-free coverage of the terrestrial surface. The HyspIRI Multi-Spectral Thermal instrument provides: 8 spectral bands from 4 to 12 microns; 60 m spatial sampling with a 600 km swath; and fully downlinked coverage of the Earth's terrestrial shallow water regions every 5 days (day/night) to provide nominally cloud-free monthly coverage. The HyspIRI mission also includes an on-board processing and direct broadcast capability, referred to as the Intelligent Payload Module (IPM), which will allow users with the appropriate antenna to download a subset of the HyspIRI data stream to a local ground station. These science and science application objectives are critical today and uniquely addressed by the combined imaging spectroscopy, thermal infrared measurements, and IPM direct broadcast capability of HyspIRI. Two key objectives are: (1) The global HyspIRI spectroscopic measurements of the terrestrial biosphere including vegetation composition and function to constrain and reduce the uncertainty in climate-carbon interactions and terrestrial biosphere feedback. (2) The global 8 band thermal measurements to provide improved constraint of fire related emissions. In this paper the current HyspIRI mission concept that has been reviewed and refined to its current level of maturity with a Data Products Symposium, Science Workshop and NASA HWorkshop is presented including traceability between the measurements and the science and science application objectives.

Green, Robert O.↗

Effect of salinity on the preservation of biomarkers in hypersaline microbial mat kerogens

Kerogen is the residue that remains after the minerals and organic compounds have been extracted from a rock using acid and organic solvents, respectively [1]. It is not only the largest organic carbon sink in the geological record but the best-preserved fraction due to its recalcitrance and ability to retain syngenetic source information during deposition [2]. Due to the complex, macromolecular, cross-linked polymeric structure of kerogen, it is difficult to analyse intact, often requiring breakdown into smaller subunits through thermal (pyrolysis) or chemical treatment (chemolysis) [3]. Saline and hypersaline environments host diverse microbial communities, which can trap and bind organic matrices such as cellular components (e.g., extracellular polymeric substances, cell wall, peptidoglycan, sheath material, etc.) with sediments, forming layered, laminated structures called microbial mats. Here we present depth profiles of two hypersaline cores through a microbial mat with differing total salinities. We looked at the composition of organic compounds released from kerogen during high temperature, hydrogen-assisted, pyrolysis to understand the effect of salinity on the composition and timing of their incorporation [4]. Field Site: The Exportadora del Sal S. A. (ESSA) saltern of Guerrero Negro (Baja California Sur, Mexico) lies within a typical sabhka environment adjacent to Laguna Ojo de Liebre. Note the spatial extent of the 13 major ponds of the ESSA system. The GN ponds thrive under well-monitored and stable conditions with steady state accretion and degradation of mat layers (around 0.5–1.4 cm per year). Organic mats typically attain up to 10 cm in thickness, depending on the location, which represents about 60 years of growth since these ponds were established. Sedimentation rate is rapid in these non-lithifying and non-mineralising mats, 1–1.5 cm per year. This environment is ideal for investigating in situ lipid preservation and diagenesis without the influence of constantly changing physical parameters. Methods: P4n5 and P5AB cores were collected as ~9 cm and 6.5 cm fresh cores, respectively, in June 2001 and September 2010, respectively from the ESSA saltern. The P4n5 core (9-9.1% salinity) was taken from Pond 4 near 5 and sub-divided into 10 layers. The P5AB core (11.2% salinity) was taken from Pond 5A near 5B and sub-divided into 8 layers. Lyophilised microbial mat powders were solvent-extracted using a modified Bligh-Dyer method and analysed by gas chromatography-mass spectrometry (GC-MS). Additionally, a subset of layers—layer 4 and layer 7—which represented active biomass and sediment-processed, respectively, were subjected to mild-acid methanolysis. The respective residues from conventional solvent extraction and acid methanolysis were loaded with a molybdenum sulfide catalyst and placed into a stainless-steel reactor. Samples were run on the hydropyrolysis set up (heated to 500 °C with 13-15 MPa of H2 pressure), extracted, and analysed by GC-MS. Results: Molecular profiles from the two ponds were significantly different and reflect the contributions of different photosynthetic and respiratory microbial communities to preserved organic matter. For example, the higher salinity P5AB core showed evidence of greater archaeal inputs (similarly observed in lab culture experiments [5]) as well as potential kerogen-bound carotenoids/carotenoid rearranged products or fragments. A hydrophobic emulsion formed during the acid methanolysis processing of layer 4 from both P4n5 and P5AB and layer 7 of P5AB. Hydropyrolysis of this hydrophobic residues released a greater abundance of polycyclic lipids relative to the pre-extracted control. Implications and Future Work: Catalytic hydropyrolysis of kerogen can rapidly generate abundant saturated pyrolysate products from the bound biomarker pool without altering the structures or stereochemistries of the products [6]. Chemical processing of solvent-extractable residues indicated that cellular matrices such as extracellular polymeric substances (EPS) may play a role in the sequestration of polycyclic lipid biomarkers such as steranes and hopanes. Their higher relative abundance compared to the pre-extracted control in the higher salinity layers and cores preliminarily aligns with this hypothesis, although further experiments are required to confirm this. It has been well-documented that EPS plays an important role in mineralisation, specifically carbonate formation, in microbial ecosystems. EPS can enhance calcium carbonate precipitation by providing diffusion-limited sites that create alkalinity gradients in response to microbial processes [7]. It has been experimentally demonstrated that salinity influences the total amount of EPS (both loosely- and tightly-bound) which increases with increasing salinity [8]. The higher salinity at Guerrero Negro switches the microbial population from filamentous Microcoleus to being dominated by Phormidium, Oscillatoria, and unicellular cyanobacteria. We demonstrated that lipid binding into kerogen via strong covalent linkages occurs at the very earliest stages of sedimentary diagenesis. We are currently investigating the influence of salinity on organic preservation through experimental and modelling approaches. Understanding how key biomarkers transform into preserved organic matter (i.e., from precursor biolipids to bound geolipids) in brine ecosystems will aid the search for organic biosignatures on other planetary bodies, especially Icy Moons and modern Mars.

hypersaline↗

A Global End-Member Approach to Derive aCDOM(440) from Near-Surface Optical Measurements

This study establishes an optical inversion scheme for deriving the absorption coefficient of colored (or chromophoric, depending on the literature) dissolved organic material (CDOM) at the 440 nm wavelength, which can be applied to global water masses with near-equal efficacy. The approach uses a ratio of diffuse attenuation coefficient spectral end members, i.e., a short and long wavelength pair. The global perspective is established by sampling "extremely" clear water plus a generalized extent in turbidity and optical properties that each span three decades of dynamic range. A unique data set was collected in oceanic, coastal, and inland waters (as shallow as 0.6 m) from the North Pacific Ocean, the Arctic Ocean, Hawaii, Japan, Puerto Rico, and the east and west coasts of the United States. The data were partitioned using subjective categorizations to define a validation quality subset of conservative water masses, i.e., the inflow and outflow of properties constrain the range in the gradient of a constituent, plus 15 subcategories of water masses that were not evolving conservatively. The dependence on subcategories was confirmed with an objective methodology based on cluster analysis techniques. The latter defined five distinct classes with validation quality data present in all classes, but which also decreased in percent composition as a function of increasing class number and optical complexity. Four different algorithms based on different validation quality end members were validated with accuracies of 1.–6.2 %, wherein pairs with the largest spectral span were most accurate. Although algorithm accuracy decreased with the inclusion of more subcategories containing non-conservative water masses, changes to the algorithm fit were small when a preponderance of subcategories were included. The high accuracy for all end-member algorithms was the result of data acquisition and data processing improvements, e.g., increased vertical sampling resolution to less than 1mm and a boundary constraint to mitigate wave focusing effects, respectively. An independent evaluation with a historical database confirmed the consistency of the algorithmic approach and its application to quality assurance, e.g., to flag data outside expected ranges, identify suspect spectra, and objectively determine the in-water extrapolation interval by converging agreement for all applicable end-member algorithms. The legacy data exhibit degraded performance (as 44 % uncertainty) due to a lack of high-quality near-surface observations, especially for clear waters wherein wave-focusing effects are problematic. The novel optical approach allows the in situ estimation of an in-water constituent in keeping with the accuracy obtained in the laboratory.

Stanford B Hooker↗

Air-Broadening of H2O as a Function of Temperature: 696 - 2163 cm(exp -1)

The temperature dependence of air-broadened halfwidths are reported for some 500 transitions in the (000)-(000) and (010)-(000) bands of H2(16)O using gas sample temperatures ranging from 241 to 388 K. These observations were obtained from infrared laboratory spectra recorded at 0.006 to 0.011 cm(exp-1) resolution with the McMath-Pierce Fourier transform spectrometer located at Kitt Peak. The experimental values of the temperature dependence exponents, eta, were grouped into eight subsets and fitted to empirical functions in a semi-global procedure. Overall, the values of eta were found to decrease with increasing rotational quantum number J. The number of measurements (over 2200) and transitions (586) involved exceeds by a large margin that of any other comparable reported study.

Toth, R. A.↗

Circulating miRNA Signature Predicts and Rescues Spaceflight Associated Health Risks

We have identified and validated a spaceflight-associated microRNA (miRNA) signature that is shared by rodents and humans in response to simulated and real spaceflight and regulates vascular damage by simulated deep space radiation. In previous studies, we identified miRNAs that are predicted to regulate rodent responses to spaceflight in low-Earth orbit. Here we have validated the expression of key spaceflight-associated miRNAs in rodents reacting to simulated spaceflight conditions (exposure to ionizing radiation combined with simulated microgravity) and in astronaut samples from the NASA Twin Study. We demonstrate that this circulating miRNA siganture is conserved between these different species and will serve as an ideal biomarker to monitor increase health risk due to exposure to the space environment. Lastly, to demonstrate their physiological relevance, we have utilized antagomirs to target a subset of miRNAs from the spaceflight associated signature to successfully rescue simulated deep space radiation-mediated damage in human 3D vascular constructs. These results provide strong evidence to further develop a miRNA based countermeasure in astronauts.

Afshin Beheshti↗

Architectural design of an Algol interpreter

The design of a syntax-directed interpreter for a subset of Algol is described. It is a conceptual design with sufficient details and completeness but as much independence of implementation as possible. The design includes a detailed description of a scanner, an analyzer described in the Floyd-Evans productions, a hash-coded symbol table, and an executor. Interpretation of sample programs is also provided to show how the interpreter functions.

Jackson, C. K.↗

Formal methods demonstration project for space applications

The Space Shuttle program is cooperating in a pilot project to apply formal methods to live requirements analysis activities. As one of the larger ongoing shuttle Change Requests (CR's), the Global Positioning System (GPS) CR involves a significant upgrade to the Shuttle's navigation capability. Shuttles are to be outfitted with GPS receivers and the primary avionics software will be enhanced to accept GPS-provided positions and integrate them into navigation calculations. Prior to implementing the CR, requirements analysts at Loral Space Information Systems, the Shuttle software contractor, must scrutinize the CR to identify and resolve any requirements issues. We describe an ongoing task of the Formal Methods Demonstration Project for Space Applications whose goal is to find an effective way to use formal methods in the GPS CR requirements analysis phase. This phase is currently under way and a small team from NASA Langley, ViGYAN Inc. and Loral is now engaged in this task. Background on the GPS CR is provided and an overview of the hardware/software architecture is presented. We outline the approach being taken to formalize the requirements, only a subset of which is being attempted. The approach features the use of the PVS specification language to model 'principal functions', which are major units of Shuttle software. Conventional state machine techniques form the basis of our approach. Given this background, we present interim results based on a snapshot of work in progress. Samples of requirements specifications rendered in PVS are offered to illustration. We walk through a specification sketch for the principal function known as GPS Receiver State processing. Results to date are summarized and feedback from Loral requirements analysts is highlighted. Preliminary data is shown comparing issues detected by the formal methods team versus those detected using existing requirements analysis methods. We conclude by discussing our plan to complete the remaining activities of this task.

Divito, Ben L.↗

The galaxy activity-interaction connection. I - Optical observations

As part of an investigation into the triggering of nuclear radio emission in interacting galaxies, new multicolor, broadband data are presented from a subset of Arp-Madore Catalog galaxies selected on the basis of their optical morphology - systems with loops and/or tails. The CTIO 4 m is used to obtain images in B, V, R, and I for the purpose of performing synthetic aperture photometry. Isophotal ellipse fitting to the more regular objects in the sample is also performed to quantify their isophotal deviation from purely elliptical morphologies. Color index maps for the galaxies are constructed, and the optical morphologies, broadband colors, and detailed photometric properties as well as the IRAS fluxes for the galaxies are discussed. These data are used to obtain a rough measure of the level of activity present in these systems.

Smith, Eric P.↗

Laser-Ablation ICP-MS Analyses of Meteoritic Metal Grains in Lunar Impact-Melt Breccias

Lunar impact-melt breccias contain metal grains from the meteorites that formed the breccias. Because the breccias contain clastic material that may derive from older breccias, metal grains from earlier impacts may be present, too. The large subset of moderately mafic (8 - 12% FeO), KREEP-rich ("LKFM") melt breccias is particularly important because: (1) these are the melt breccias most likely to have been produced in basin-forming impacts, (2) it is from these breccias that many of the approx. 3.9 Gyr ages that are so common in lunar samples derive, (3) the breccias contain large proportions of FeNi metal, more than 1% in some types of Apollo 16 breccias, and (4) the metal potentially provides information about the impactors causing the apparent cataclysm at 3.9 Gyr.

Korotev, R. L.↗

Developing Natural Language Processing and Supervised Learning Techniques to Classify Mars Tasks

As NASA's Human Research Program (HRP) prepares for long-duration Mars missions, understanding astronaut tasks is crucial. This study, conducted at NASA Glenn Research Center (GRC), employed Natural Language Processing (NLP) and machine learning techniques to analyze and classify Mars tasks. A list of 1,058 Mars tasks was provided by HRP experts including binary labeling of 18 Human System Task Categories (HSTCs). We developed an NLP model using Google's BERT language model to capture the semantic and syntactic nuances of these tasks. Supervised training was initially applied to a subset of the NLP-analyzed tasks to assess the model's effectiveness in classifying the remaining tasks. Incorporating HSTC descriptions significantly enhanced the classification accuracy for 9 out of the 18 HSTCs and reduced training time. To address the issue of severe class imbalance in the HSTC data, we introduced innovative weighting and sampling techniques for data augmentation. We then fine-tune BERT to implement a pairwise relatedness scoring method, allowing us to cluster tasks based on their relatedness and similarity, getting a step closer to labeling the tasks without supervision. In this presentation we guide you through data preprocessing, deciphering key syntax components using BERT, and performing supervised classification of the Mars tasks. This work showcases the potential use of advanced NLP techniques to analyze Mars missions to be incorporated into various crew health and performance analyses.

GenAI↗