Search NASA⌕ Search

SEARCH · Search NASA

Results for “text extraction”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS1983

DOCU-TEXT: A tool before the data dictionary

DOCU-TEXT, a proprietary software package that aids in the production of documentation for a data processing organization and can be installed and operated only on IBM computers is discussed. In organizing information that ultimately will reside in a data dictionary, DOCU-TEXT proved to be a useful documentation tool in extracting information from existing production jobs, procedure libraries, system catalogs, control data sets and related files. DOCU-TEXT reads these files to derive data that is useful at the system level. The output of DOCU-TEXT is a series of user selectable reports. These reports can reflect the interactions within a single job stream, a complete system, or all the systems in an installation. Any single report, or group of reports, can be generated in an independent documentation pass.

Carter, B.↗
NASA NTRS1994

Database tomography for commercial application

Database tomography is a method for extracting themes and their relationships from text. The algorithms, employed begin with word frequency and word proximity analysis and build upon these results. When the word 'database' is used, think of medical or police records, patents, journals, or papers, etc. (any text information that can be computer stored). Database tomography features a full text, user interactive technique enabling the user to identify areas of interest, establish relationships, and map trends for a deeper understanding of an area of interest. Database tomography concepts and applications have been reported in journals and presented at conferences. One important feature of the database tomography algorithm is that it can be used on a database of any size, and will facilitate the users ability to understand the volume of content therein. While employing the process to identify research opportunities it became obvious that this promising technology has potential applications for business, science, engineering, law, and academe. Examples include evaluating marketing trends, strategies, relationships and associations. Also, the database tomography process would be a powerful component in the area of competitive intelligence, national security intelligence and patent analysis. User interests and involvement cannot be overemphasized.

Kostoff, Ronald N.↗
DOE OSTI2024

Protein–Protein Interaction Networks Derived from Classical and Machine Learning-Based Natural Language Processing Tools

The study of protein-protein interactions (PPIs) provides insight into various biological mechanisms, including the binding of antibodies to antigens, enzymes to inhibitors or promoters, and receptors to ligands. Recent studies of PPIs have led to significant biological breakthroughs. For example, the study of PPIs involved in the human:SARS-CoV-2 viral infection mechanism aided in the development of the SARS-CoV-2 vaccines. Though several databases exist for the manual curation of PPI networks, text mining methods have been routinely demonstrated as useful alternatives for newly studied or understudied species where databases are incomplete. Here, the relationship extraction (RE) performance of several open-source classical text processing, machine learning (ML)-based natural language processing (NLP), and large language model (LLM)-based NLP tools were compared. Overall, our results indicated that networks derived from classical methods tend to have high true positive rates at the expense of having overconnected-networks, ML-based NLP methods have lower true positive rates but networks with the closest structures to the target network, and LLM-based NLP methods tend to exist in-between the two other approaches, with variable performances. Finally, the selection of a specific NLP approach should be tied to the needs of a study and text availability, as models varied in performance due to the amount of text provided.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Deformable phrase level attention: A flexible approach for improving AI based medical coding

Objective: Improving the AI-driven automated medical encoding of clinical text plays a vital role in gathering information on the occurrence of diseases to improve population-level health. This work presents a novel attention mechanism designed to enhance text classification models and ensure appropriate classification of medical concepts in unstructured electronic health records. Materials and Methods: We developed a deformable, phrase-level attention mechanism to identify important lexical word-level and contextual phrase-level information from clinical text documents. We evaluated conventional and transformer-based deep learning models that we extended with our attention mechanism on the extraction of critical cancer information (e.g., site, subsite, laterality, histology, behavior) from 629,908 electronic pathology reports and on the automated medical encoding of 52,722 hospital discharge summaries. Results: Transformer-based models with the deformable, phrase-level attention mechanism achieved the best performance on the extraction of critical cancer information from pathology reports. Conventional- and transformer-based models show similar or better performance than their baseline counterparts on the automated medical encoding of clinical documents. Discussion: The addition of phrase-level information allowed models extended with our proposed method to outperform standard word-level attention. Our method showed favorable properties for the real-world application in terms of model robustness and phenotyping. These results indicate that our method is promising for automated data harmonization for common data models. Conclusion: This work proposes a novel deformable, phrase-level attention mechanism that enhances text classification models in the extraction of medical concepts from clinical text documents. We demonstrate strong performances on two clinical text datasets and showcase real-world deployability of our method.

Automated medical encoding↗
DOE OSTI2024

PhysBERT: A text embedding model for physics scientific literature

The specialized language and complex concepts in physics pose significant challenges for information extraction through Natural Language Processing (NLP). Central to effective NLP applications is the text embedding model, which converts text into dense vector representations for efficient information retrieval and semantic analysis. In this work, we introduce PhysBERT, the first physics-specific text embedding model. Pre-trained on a curated corpus of 1.2 × 106 arXiv physics papers and fine-tuned with supervised data, PhysBERT outperforms leading general-purpose models on physics-specific tasks, including the effectiveness in fine-tuning for specific physics subdomains.

Hellert, Thorsten (ORCID:0000000227970926)↗
DOE OSTI2025

Rapid Adaptation of Chemical Named Entity Recognition Using Few-Shot Learning and LLM Distillation

Named entity recognition (NER) has been widely used in chemical text mining for the automatic identification and extraction of chemical entities. However, existing chemical NER systems primarily focus on scenarios with abundant training data, requiring significant human effort on annotations. This poses challenges for applications in the chemical field, such as catalysis, where many advancements have traditionally relied on trial-and-error investigations and incremental adjustment of variables. This hinders catalysis science and technology progress in addressing emerging energy and environmental crises. In this work, we propose a few-shot NER model that can quickly adapt to extract new types of chemical entities by using only a limited number of annotated examples. Our model employs a metric-learning approach to transfer entity similarity knowledge from high-resource chemical domains (with abundant annotations) to enable effective entity recognition in low-resource specialized domains (limited annotation). We validate the effectiveness of our model on a few-shot chemical NER benchmark built based on six existing chemical NER data sets. Experiments show that the proposed few-shot NER model can achieve reasonable performance with only 5 examples per entity type and shows consistent improvement as the number of examples increases. Furthermore, we demonstrate how the proposed model can be trained with large language model (LLM) annotated data, opening a new pathway for rapid adaptation of NER systems. Furthermore, our approach leverages the knowledge broadness of large language models for chemistry while distilling this knowledge into a lightweight model suitable for efficient and in-house use.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1990

Solar Astronomy Data Base: Packaged Information on Diskette

In its role as a library, the National Geophysical Data Center has transferred to diskette a collection of small, digital files of routinely measured solar indices for use on an IBM-compatible desktop computer. Recording these observations on diskette allows the distribution of specialized information to researchers with a wide range of expertise in computer science and solar astronomy. Every data set was made self-contained by including formats, extraction utilities, and plain-language descriptive text. Moreover, for several archives, two versions of the observations are provided - one suitable for display, the other for analysis with popular software packages. Since the files contain no control characters, each one can be modified with any text editor.

Mckinnon, John A.↗
NASA NTRS2022

Event Report for The Ethical Artificial Intelligence Quantification Workshop

Artificial Intelligence (AI) is a powerful emerging technology area which requires special attention to using it ethically. AI ethics is still an emerging field, and the partners for this workshop and report seek to move AI ethics discussion ahead by experimenting with ways to measure AI ethics criteria. The following document describes the outcomes and learnings from The Ethical Artificial Intelligence Quantification Workshop held at the National Institute for Aerospace (NIA), Hampton, Virginia on May 12th, 2022. The purpose of the workshop was for participants to evaluate and experiment-with the methodology and process presented by AIEthics.World in cooperation with Intel Corporation. The meeting participants learned about the Ethical AI Certification and Maturity Model™ and applied the methodology to selected notional AI systems. The workshop facilitated the evaluation of the maturity of the AI system according to ethical considerations relevant to NASA, NIA and other participants. The workshop consisted of three main phases. The first phase focused on understanding and summarizing NASA’s ethical approaches, mission and values based on published documentation, discussions and individual insights & opinions of participants. This information was prioritized, weighted, ordered, and quantified in phase two, to formulate an alignment between human values (ethics) and their applicability to AI systems during all lifecycle phases. The first two phases were summarized as a form of ethical genealogy for artificial intelligence, specific to NASA’s ethical approaches. In the third and last phase of the workshop the participants evaluated notional examples of artificial intelligence to qualify and quantify its ability to adhere to the organizational ethics approaches, using the Ethical AI Certification and Maturity Model™. The workshop uses the concept of genealogy, in the traditional sense: the study and traceability of lines of ancestors in the process of evolutionary development from earlier forms. However, as it is applied to an Ethical AI definition, it is providing the insights to the necessary and mandatory traceability of content, data, metrics, telemetry, elements, and structures which are used in the AI’s lifecycle to foster and measure AI ethics in all steps of its lifecycle. The Ethical Artificial Intelligence Quantification Workshop provided NASA with the opportunity to apply the Ethical AI Certification and Maturity Model™, in combination with existing and well-known decision-making and quality control methods to identify the metrics and measurements for an Ethical AI and assess its ethical condition and quality aligned with NASA ethics approaches. The result of the workshop is the capacity for NASA to apply the maturity model assessment to its AI Systems as desired and if necessary, publish the ability of these AI Systems to adhere to the organizational ethical goals. AI ethics frameworks need to be customized for each application domain, for example, individual NASA Mission Directorates. General principles that work in one area such as AI/Machine Learning-based text analysis (the ethics of information-extraction) may need to be adapted for another such as sense-and-avoid decision-making in a flight environment. The workshop was conducted among approximately twenty NASA subject matter experts, so the elements noted above should be considered examples, not definitive NASA ethical AI principles, genealogy, etc. Generating a definitive AI ethics framework for an organization as diverse as NASA would require far more discussion, debate, review, etc. However, the workshop provided valuable insight into mechanisms and processes for quantifying AI ethical qualities.

Artificial Intelligence↗
NASA NTRSDate not supplied

Digitizing Named Entities Found Within Letters of Agreement

Letters of Agreement (LOAs) are text-based air traffic control documents that contain procedures and actions agreed upon by the different parties, typically two or more FAA facilities, that are subject to an agreement. The documents contain among other things generic constraints, which are explicit and implicit combinations of procedures that limit a flight’s trajectory and affects pilot actions. For example, a controller may be required, to assign a specific altitude to an aircraft crossing the boundary between two airspaces. Although LOA generic constraints directly impact the trajectory of an aircraft, they are not currently available in a digital form that can be used for (or directly ingested into automated) flight planning. Instead, the constraints are manually input into an onboard or ground based system. LOA documents are primarily stored at a controlling facility and the generic constraints are implemented by experienced air traffic controllers and pilots primarily using voice instructions. This increases the workload of the controllers, likelihood of error (e.g., due to noisy communication) and makes it impractical for implementation with unmanned aircraft. Therefore, steps must be taken to make existing constraints machine interpretable to enable e.g., automated handoffs which in turn would reduce controller workload. With recent advances in natural language processing, especially the rise in digitization of text documents (e.g., medical documents) and automated extraction of information therein, it is now possible to extract flight specific constraints from LOAs. The goal of this work is to digitize named entities through a combination of natural language processing tasks: named entity disambiguation, toponym resolution, and numeric parsing to extract general constraint components contained within LOAs, herein referred to as Entity Enhancement (EE). Starting with a small list of named entities (e.g., ARTCC, Tower, Altitude and Speed), EE can extract the named entities while simultaneously converting the string-based output into a digital format using an ensemble of processes like rule-based gazetteers and syntactic-lexical patterns. The digital format contains a diverse set of information based on the entity label in question, ranging from standardized facility names to units of measure (e.g., feet) and other numeric information. Upon validating our approach using a truth dataset, we show an overall F1-Score of 0.71 for the extraction process. Looking beyond entity enhancement, we are also working towards the goal of completely digitizing the general constraints by performing EE and fitting them into a standardized exchange model (XM) such as the Aeronautical Information Exchange Model (AIXM). This will allow for easy distribution and dissemination of LOA constraints to air users, better searchability within documents, and enable ingestion into automated flight planning. Finally, we show a preliminary version of the proposed XM architecture and demonstrate how the model can be populated from the EE output.

Stephen S. B. Clarke↗
DOE OSTI2025

A Model Based Approach to Extract Health Information from Textual Data

In current nuclear power plants (NPPs) a large amount of condition-based data is being generated and stored to assess and monitor component health and performance. The format of this data can be either numeric (e.g., pump vibration data) or textual (e.g., condition report which assess component health). While assessing component health from numeric data can be performed with a large variety of methods, the extraction of information from textual data still remains a challenge. Natural language processing (NLP) methods are starting to be deployed in current NPPs mainly to filter out incident reports (IRs) that are not safety related by employing supervised machine learning methods. However, these methods do not really provide the quantitative information that might be contained in IRs. This paper presents an approach to extract information from textual data (e.g., from IRs, maintenance reports) that is based on NLP data analytics methods coupled with model-based system engineer (MBSE) models. NLP methods are employed to perform syntactic and semantic analyses. Syntactic analysis analyzes the grammatical structure of a sentence; such analysis includes: part of speech (POS) tagging (i.e., identification of grammatic elements of each string - e.g., nouns, verbs), named entity recognition (i.e., identification of text entities - e.g., names, dates, events), and relation extraction (e.g., coreference resolution). On the other hand, semantic analysis is designed to analyze the logic structure of a sentence. Through a specific set of rules, our methods can identify whether a sentence contains health information of a component (e.g., degraded performance, anomaly behavior) or the causal relationship between two events (i.e., a cause-effect pair). An innovative element of our approach is that semantic analysis relies on MBSE models to identify links between textual elements. MBSE are diagrams designed to represent system and component dependencies (from both a form and functional point of view). In our approach, MBSE models emulate system engineer knowledge about component/system architecture. This paper presents in detail how the integration of NLP methods and MBSE models is performed. Few analysis examples focusing on centrifugal pumps are presented.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Machine learning and deep learning tools for the automated capture of cancer surveillance data

The National Cancer Institute and the Department of Energy strategic partnership applies advanced computing and predictive machine learning and deep learning models to automate the capture of information from unstructured clinical text for inclusion in cancer registries. Applications include extraction of key data elements from pathology reports, determination of whether a pathology or radiology report is related to cancer, extraction of relevant biomarker information, and identification of recurrence. With the growing complexity of cancer diagnosis and treatment, capturing essential information with purely manual methods is increasingly difficult. These new methods for applying advanced computational capabilities to automate data extraction represent an opportunity to close critical information gaps and create a nimble, flexible platform on which new information sources, such as genomics, can be added. This will ultimately provide a deeper understanding of the drivers of cancer and outcomes in the population and increase the timeliness of reporting. These advances will enable better understanding of how real-world patients are treated and the outcomes associated with those treatments in the context of our complex medical and social environment.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Ocpp 2.0.1. Interim Kpi Calculator

The project is split into four pieces. The first is a raw OCPP log parser. The second is a file splitter. The third is a message parser. The final piece is the Interim KPI calculator. The OCPP log parser was created from two different formats of raw OCPP 2.0.1 data. Its intended purpose is to extract device IDs and OCPP event messages from nontabular text logs. The parser looks for specific substrings in the logs to identify which of the two "standards" it should select from. The KPI generator does not perform any of its calculations in parallel. Instead, we opt for a naive batching approach. The splitter takes the file generated from the parser and creates many smaller files for each of the device IDs in the dataset. This allows the pandas queries in the log formatter to be iterate over a significantly smaller slice of data, increasing performance significantly. The message parser step takes messages from each of the files (containing distinct device IDs) and breaks the message out into pieces. The final result is a file with different columns specifying different attributes of the JSON message. The file is an aggregation of all different devices. This is the most complex portion of the code. The KPI calculator takes the parsed messages, as a single file, and calculates the KPI from that data. An excel file is produced with four sheets. These contain the metrics for Session Success, Charge Start Success, Charge End Success, and Charge Start Time. It includes the metrics for the different equations in the Interim KPI Implementation Guide as well as a weighted sum of the different equations for each KPI (excluding Charge End Success and Charge Start Time).

Quinn, Casey↗
NASA NTRS2018

GMI-IPS: Python Processing Software for Aircraft Campaigns

NASA's Atmospheric Tomography Mission (ATom) seeks to understand the impact of anthropogenic air pollution on gases in the Earth's atmosphere. Four flight campaigns are being deployed on a seasonal basis to establish a continuous global-scale data set intended to improve the representation of chemically reactive gases in global atmospheric chemistry models. The Global Modeling Initiative (GMI), is creating chemical transport simulations on a global scale for each of the ATom flight campaigns. To meet the computational demands required to translate the GMI simulation data to grids associated with the flights from the ATom campaigns, the GMI ICARTT Processing Software (GMI-IPS) has been developed and is providing key functionality for data processing and analysis in this ongoing effort. The GMI-IPS is written in Python and provides computational kernels for data interpolation and visualization tasks on GMI simulation data. A key feature of the GMI-IPS, is its ability to read ICARTT files, a text-based file format for airborne instrument data, and extract the required flight information that defines regional and temporal grid parameters associated with an ATom flight. Perhaps most importantly, the GMI-IPS creates ICARTT files containing GMI simulated data, which are used in collaboration with ATom instrument teams and other modeling groups. The initial main task of the GMI-IPS is to interpolate GMI model data to the finer temporal resolution (1-10 seconds) of a given flight. The model data includes basic fields such as temperature and pressure, but the main focus of this effort is to provide species concentrations of chemical gases for ATom flights. The software, which uses parallel computation techniques for data intensive tasks, linearly interpolates each of the model fields to the time resolution of the flight. The temporally interpolated data is then saved to disk, and is used to create additional derived quantities. In order to translate the GMI model data to the spatial grid of the flight path as defined by the pressure, latitude, and longitude points at each flight time record, a weighted average is then calculated from the nearest neighbors in two dimensions (latitude, longitude). Using SciPya's Regular Grid Interpolator, interpolation functions are generated for the GMI model grid and the calculated weighted averages. The flight path points are then extracted from the ATom ICARTT instrument file, and are sent to the multi-dimensional interpolating functions to generate GMI field quantities along the spatial path of the flight. The interpolated field quantities are then written to a ICARTT data file, which is stored for further manipulation. The GMI-IPS is aware of a generic ATom ICARTT header format, containing basic information for all flight campaigns. The GMI-IPS includes logic to edit metadata for the derived field quantities, as well as modify the generic header data such as processing dates and associated instrument files. The ICARTT interpolated data is then appended to the modified header data, and the ICARTT processing is complete for the given flight and ready for collaboration. The output ICARTT data adheres to the ICARTT file format standards V1.1. The visualization component of the GMI-IPS uses Matplotlib extensively and has several functions ranging in complexity. First, it creates a model background curtain for the flight (time versus model eta levels) with the interpolated flight data superimposed on the curtain. Secondly, it creates a time-series plot of the interpolated flight data. Lastly, the visualization component creates averaged 2D model slices (longitude versus latitude) with overlaid flight track circles at key pressure levels. The GMI-IPS consists of a handful of classes and supporting functionality that have been generalized to be compatible with any ICARTT file that adheres to the base class definition. The base class represents a generic ICARTT entry, only defining a single time entry and 3D spatial positioning parameters. Other classes inherit from this base class; several classes for input ICARTT instrument files, which contain the necessary flight positioning information as a basis for data processing, as well as other classes for output ICARTT files, which contain the interpolated model data. Utility classes provide functionality for routine procedures such as: comparing field names among ICARTT files, reading ICARTT entries from a data file and storing them in data structures, and returning a reduced spatial grid based on a collection of ICARTT entries. Although the GMI-IPS is compatible with GMI model data, it can be adapted with reasonable effort for any simulation that creates Hierarchical Data Format (HDF) files. The same can be said of its adaptability to ICARTT files outside of the context of the ATom mission. The GMI-IPS contains just under 30,000 lines of code, eight classes, and a dozen drivers and utility programs. It is maintained with GIT source code management and has been used to deliver processed GMI model data for the ATom campaigns that have taken place to date.

Damon, M. R.↗
DOE OSTI2024

Search for ${\text {Z}{}{}} {\text {Z}{}{}} $ and ${\text {Z}{}{}} {\text {H}{}{}} $ production in the ${\text {b}{}{}} {\bar{{\text {b}{}{}}}{}{}} {\text {b}{}{}} {\bar{{\text {b}{}{}}}{}{}} $ final state using proton-proton collisions at $\sqrt{s}=13\,\text {Te}\hspace{-.08em}\text {V} $

A search for ${\text {Z}{}{}} {\text {Z}{}{}} $ and ${\text {Z}{}{}} {\text {H}{}{}} $ production in the ${\text {b}{}{}} {\bar{{\text {b}{}{}}}{}{}} {\text {b}{}{}} {\bar{{\text {b}{}{}}}{}{}} $ final state is presented, where H is the standard model (SM) Higgs boson. The search uses an event sample of proton-proton collisions corresponding to an integrated luminosity of 133$\,\text {fb}^{-1}$ collected at a center-of-mass energy of 13$\,\text {Te}\hspace{-.08em}\text {V}$ with the CMS detector at the CERN LHC. The analysis introduces several novel techniques for deriving and validating a multi-dimensional background model based on control samples in data. A multiclass multivariate classifier customized for the ${\text {b}{}{}} {\bar{{\text {b}{}{}}}{}{}} {\text {b}{}{}} {\bar{{\text {b}{}{}}}{}{}} $ final state is developed to derive the background model and extract the signal. The data are found to be consistent, within uncertainties, with the SM predictions. The observed (expected) upper limits at 95% confidence level are found to be 3.8 (3.8) and 5.0 (2.9) times the SM prediction for the ${\text {Z}{}{}} {\text {Z}{}{}} $ and ${\text {Z}{}{}} {\text {H}{}{}} $ production cross sections, respectively.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

Scaling open-weight large language models for hydropower regulatory information extraction: A systematic analysis

Information extraction from regulatory and technical documents using large language models (LLMs) involves practical trade-offs between extraction quality and computational cost. We evaluate eight open-weight LLMs spanning 0.6B–70B parameters on hydropower licensing documents and report deployment-oriented evidence under a unified extraction schema and evaluation protocol. Across the model set, we observe clear scale-dependent trends in both baseline extraction quality and the effectiveness of reflective reasoning (self-checking) under our fixed-prompt, no-augmentation setting. Mid-scale models often provide a favorable balance of accuracy and efficiency, whereas the smallest models show limited or inconsistent gains from the reasoning variants tested. Larger models achieve the highest overall F1 scores but incur substantially greater compute and infrastructure requirements. We further find that reliability failure modes can distort conventional metrics in this domain: in particular, high recall can coincide with systematic extraction errors when models fabricate values for fields that are absent from the source text, underscoring the importance of conservative null handling and evidence-grounded evaluation. Overall, our study provides a reproducible resource–performance comparison for open-weight LLM-based extraction in hydropower regulatory documentation and offers practical guidance for model selection under different deployment constraints.

Evaluation protocol↗
NASA NTRS1978

A transportable system for management and exchange of programs and other text

A system is presented for exchanging information on tape, that allows information about the data to be included with the data. The system is designed for portability, but requires a few simple machine dependent modules. These modules are available for a variety of machines, and a bootstrapping procedure is provided. The system allows content selected reading of the tape, and a simple text editing facility is provided. Although the system recognizes 30 commands, information may be extracted from the tape by using as few as three commands. In addition to its use for information exchange, the system is expected to find use in maintaining large libraries of text.

Snyder, W. V.↗
NASA NTRSDate not supplied

What Went Wrong: A Survey of Wildfire UAS Mishaps through Named Entity Recognition

Increasingly, unmanned aircraft systems (UAS) are being applied to wildfire incidents for tasks such as mapping, aerial ignition, and delivery. As a result, aviation incident reporting systems for wildfires are beginning to accumulate data related to UAS mishaps in wildfire response. In this research, we apply state-of-the-art natural language processing (NLP) techniques to develop a custom Named Entity Recognition (NER) model which extracts entities relevant to safety analysts. The custom NER model is built by fine-tuning an existing Bidirectional Encoder Representations from Transformers (BERT) model, resulting in a generalizable NER model that can extract engineering relevant entities including failure modes, causes, effects, control processes, and recommendations from failure-relevant text. This model performs passably, with a weighted average f1 score of 0.33 across entity types, indicating more labeled training data is needed. Extracted entities are used to form a Failure Modes and Effects Analysis (FMEA)-style survey of wildfire UAS mishaps reported using the SAFECOM system. Similar mishaps are manually clustered and reported as single rows within an FMEA. Foreach cluster, we compute frequency, severity, and overall riskin accordance with FAA standards. This methodology can beapplied as part of a broader safety management system totrack trends in mishaps (e.g., likelihood, severity) and discoverknowledge (e.g., causes, effects) that can be utilized to improvesafety outcomes and system performance.

Machine Learning↗