Visual data mining for quantized, spatial data
Explore the source record for details and available documents.
SEARCH · Engineering Papers
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.
Explore the source record for details and available documents.
In a previous study, multiple regression techniques were applied to Flight Operations Quality Assurance-derived data to develop parsimonious model(s) for fuel consumption on the Boeing 757 airplane. The present study examined several data mining algorithms, including neural networks, on the fuel consumption problem and compared them to the multiple regression results obtained earlier. Using regression methods, parsimonious models were obtained that explained approximately 85% of the variation in fuel flow. In general data mining methods were more effective in predicting fuel consumption. Classification and Regression Tree methods reported correlation coefficients of .91 to .92, and General Linear Models and Multilayer Perceptron neural networks reported correlation coefficients of about .99. These data mining models show great promise for use in further examining large FOQA databases for operational and safety improvements.
Ensemble Data Mining Methods, also known as Committee Methods or Model Combiners, are machine learning methods that leverage the power of multiple models to achieve better prediction accuracy than any of the individual models could on their own. The basic goal when designing an ensemble is the same as when establishing a committee of people: each member of the committee should be as competent as possible, but the members should be complementary to one another. If the members are not complementary, Le., if they always agree, then the committee is unnecessary---any one member is sufficient. If the members are complementary, then when one or a few members make an error, the probability is high that the remaining members can correct this error. Research in ensemble methods has largely revolved around designing ensembles consisting of competent yet complementary models.
A data mining technique called 'k-means clustering' can be used to group winds at the NWTC into 4 major clusters. The frequency of some winds in the clusters is correlated with regional pressure gradients and climate indices. The technique could also be applied to wind resource assessment and selecting scenarios for flow modeling.
We present a data mining framework for the analysis and discovery of anomalies in high-dimensional time series of sensor measurements that would be found in an Integrated System Health Monitoring system. We specifically treat the problem of discovering anomalous features in the time series that may be indicative of a system anomaly, or in the case of a manned system, an anomaly due to the human. Identification of these anomalies is crucial to building stable, reusable, and cost-efficient systems. The framework consists of an analysis platform and new algorithms that can scale to thousands of sensor streams to discovers temporal anomalies. We discuss the mathematical framework that underlies the system and also describe in detail how this framework is general enough to encompass both discrete and continuous sensor measurements. We also describe a new set of data mining algorithms based on kernel methods and hidden Markov models that allow for the rapid assimilation, analysis, and discovery of system anomalies. We then describe the performance of the system on a real-world problem in the aircraft domain where we analyze the cockpit data from aircraft as well as data from the aircraft propulsion, control, and guidance systems. These data are discrete and continuous sensor measurements and are dealt with seamlessly in order to discover anomalous flights. We conclude with recommendations that describe the tradeoffs in building an integrated scalable platform for robust anomaly detection in ISHM applications.
Data mining is defined as the discovery of useful, possibly unexpected, patterns and relationships in data using statistical and non-statistical techniques in order to develop schemes for decision and policy making. Data mining can be used to discover the sources and causes of problems in complex systems. In addition, data mining can support simulation strategies by finding the different constants and parameters to be used in the development of simulation models. This paper introduces a framework for data mining and its application to complex problems. To further explain some of the concepts outlined in this paper, the potential application to the NASA Shuttle Reinforced Carbon-Carbon structures and genetic programming is used as an illustration.
The Protein Data Bank (PDB) was established as the 1st open access digital data resource in biology and medicine. Today, the PDB archive contains over 140,000 atomic‐level biomolecular structures determined by crystallography, NMR spectroscopy, and 3D electron microscopy. About 20 new structures are deposited daily. PDB data is used in basic and applied research, patent applications, discovery of lifesaving drugs, product innovation, and education. About 80% of all PDB structures deposited have a “primary citation,” which is the first paper to describe the molecule and its structure and function. In turn, these primary citations have contributed to more than 1 million publications. An initial study by Clarivate Analytics on the Nucleic Acids Research publication demonstrated that PDB data drive high‐impact research in diverse scientific fields. To build on this study, we used Python programming to perform several data mining studies to understand the impact of these structures in fundamental biology, biomedicine, and energy in the scientific literature. The scientific literature database, Web of Science, tracks scientific article metrics such as the number of times cited, subject category, year published, funding source, and keywords. Web of Science data for PDB primary citations was compiled as used as the data set. Two main investigations were performed. In the first, we examined the number of citations per article (data overall and by individual subject categories) to develop “top 10 cited structures” lists and analyzed for trends. In the second investigation, we extended our data mining analysis to consider for each main subject category, the papers with the most citations and the structures they cite, and the most frequently occuring keywords. We have observed different aspects of biological data utility in various subject areas, at times in ways unanticipated from the structures' primary citation. This research demonstrates the value of biological data from the PDB by showing the diverse ways in which the data are being used. Support or Funding Information RCSB PDB is funded by a grant (DBI‐1338415) from the National Science Foundation, the National Institutes of Health, and the US Department of Energy. This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .
The NASA Occupant Protection Group is charged with ensuring crewmembers are protected during all dynamic phases of spaceflight. Previous work with outside experts has led to the development of a definition of acceptable risk (DAR) for space capsule vehicles. The DAR defines allowable probability rates for various categories of injuries. An important question is how to validate these probabilities for a given vehicle. One approach is to impact test human volunteers under projected nominal landing loads. The main drawback is the large number of subject tests required to attain a reasonable level of confidence that the injury probability rates would meet those outlined in the DAR. An alternative is to mine existing databases containing human responses to impact. Testing an anthropomorphic test device (ATD) at the same human‐exposure levels could yield a range of ATD responses that would meet DAR. As one aspect of future vehicle validation, the ATD could be tested in the vehicle's seat and suit configuration at nominal landing loads and compared with the ATD responses supported by the human data set. This approach could reduce the number of human‐volunteer tests NASA would need to conduct to validate that a vehicle meets occupant protection standards. METHODS: The U.S. Air Force has recorded hundreds of human responses to frontal, lateral, and spinal impacts at many acceleration levels and pulse durations. All of this data are stored on the Collaborative Biomechanics Data Network (CBDN), which is maintained by the Wright Patterson Air Force Base (WPAFB). The test device for human occupant restraint (THOR) ATD was impact tested on WPAFB's horizontal impulse accelerator (HIA) matching human‐volunteer exposures on the HIA to 5 frontal and 3 spinal loading conditions. No human injuries occurred as a result of these impact conditions. Peak THOR response variables for neck axial tension and compression, and thoracic‐spine axial compression were collected. Maximal chest deflection was determined from motion capture video of the impact test. HIC‐ 15 and BRIC were calculated from head acceleration responses. Given the number of human subjects for each test condition a confidence interval of injury probability will be obtained. RESULTS: Results will be discussed in terms of injury‐risk probability estimates based on the human data set evaluated. Also, gaps in the data set will be identified. These gaps could be one of two types. One is areas where additional THOR testing would increase the comparable human data set, thereby improving confidence in the injury probability rate. The other is where additional human testing would assist in obtaining information on other acceleration levels or directions. DISCUSSION: The historical human data showed validity of the THOR ATD for supplemental testing. The historical human data are limited in scope, however. Further data are needed to characterize the effects of sex, age, anthropometry, and deconditioning due to spaceflight on risk of injury
In previous papers we've shown how a well known data compression algorithm called Entropy-constrained Vector Quantization ( can be modified to reduce the size and complexity of very large, satellite data sets. In this paper, we descuss how to visualize and understand the content of such reduced data sets.
No abstract available
Purpose: To quality assure a large quantity of retrospective treatment cases for treatment performances by randomly sampling is inefficient. Here we provide a method to efficiently monitor and investigate the QA of SBRT workflow over Mosaiq. Methods: The code developed with Microsoft SQL Server Management Studio 2008R2 and VBA was used for retrieving and sorting data from Mosaiq (version 2.3–2.6 during 2012–2015). SBRT patients were filtered by fractional dose over 350cGy and total fraction number less than 6, which SBRT prescriptions were defined. The quality assurance on the SBRT workflow was focused on the treatment deliveries such as patient positioning setup, CBCT indicated offsets and couch shifted corrections. The treatment delivery were done by Varian Truebeam systems and the record/verify by Mosaiq. Results: Total 82 SBRT patients corresponding to 103 courses and 854 CBCT images were found by the retrieval query. Most centers record daily pre-treatment (Pre-Tx: before treatment shift) image-guided shifts along treatment course for inter-fraction motion record, and it is useful to also verify it with post-treatment imaging (Post-Tx: after treatment CBCT verification) to verify intra-fraction motion. Analyzing the details of daily recorded shifts can reveals the information of patient-setup and staff’s record/verify behaviors. 3 examples were provided as solid evidences and on-going rectification for preventing future mistakes. Conclusions: The report gave feasible examples for inspector to verify a large amount of data during site investigation. This program can also be extended to a scheduled data mining with software to periodical analyze the timely records in Mosaiq, for example, a various control charts for different QA purposes. As the current trend of automation in radiation therapy field, the data mining would be a necessary tool in the future, just as the automatic plan quality evaluation has been under development in Eclipse.
The mining of data streams for online condition monitoring is a challenging task in several domains including (electric) power grid system, intelligent manufacturing, and consumer science. Considering a power grid application in which thousands of sensors, called the phasor measurement units, are deployed on the power grid network to continuously collect streams of digital data for real-time situational awareness and system management. Depending on design, each sensor could stream between ten and sixty data samples per second. The myriad of sensory data captured could convey deeper insights about sequence of events in real-time and before major damages are done. However, the timely processing and analysis of these high-velocity and high-volume data streams is a challenge. Hence, a new data processing and transformation approach, based on the concept of control charts, for representing sequence of data streams from sensors is proposed. In addition, an application of the proposed approach for enhancing data mining tasks such as clustering using real-world power grid data streams is presented. The results indicate that the proposed approach is very efficient for data streams storage and manipulation.
In this paper, we report a data-mining investigation for the search of topological insulators by examining individual electronic structures for over 60,000 materials. Using a data-mining algorithm, we survey changes in band inversion with and without spin-orbit coupling by screening the calculated electronic band structure for a small gap and a change concavity at high-symmetry points. Overall, we were able to identify a number of topological candidates with varying structures and composition. Lastly, our overall goal is expand the realm of predictive theory into the determination of new and exotic complex materials through the data mining of electronic structure.
Integrated computational materials engineering (ICME) methods combining CALPHAD with process-based simulations can produce rich, high-dimensional data for alloy and process design. In ICME methods for metallurgical applications, the visualization and interpretation of such high-dimensional data has previously been through heat maps represented in 2 or 3 dimensions. While such an approach is ideal when one variable is varied at a time, in the case of high-dimensional data with multiple variables varied simultaneously, as is the case in additive manufacturing, interpreting the trends through two- or three-dimensional heat maps becomes challenging. Here, we propose a strategy of mixed visual data mining and quantitative analysis for high-dimensional metallurgical and process data using high-throughput thermodynamic calculations. Two case studies show the application of the proposed approach. The first case study investigated the effects of feedstock chemistry on the δ ferrite formation in 316L stainless steel powders used for binder jet additive manufacturing. The second case study linked Scheil–Gulliver calculations to a process model for dissimilar joining of aluminum alloys 5356 and 6111 during laser hot-wire additive manufacturing. Both cases contained thousands of calculated data points, showcasing the utility of visual data analysis through parallel coordinate plotting, Pearson correlation coefficient matrices, and scatter matrices compared to traditional process maps. These visualization techniques can be extended to many additive manufacturing problems to capture process–structure–property relationships for additively manufactured components.
No abstract available
Air traffic service providers have to make decisions regarding changes to air traffic flow in the event of major weather disturbances and traffic congestions to maintain safety of the system. The behavior of the air traffic management system will be more predictable if consistent decisions are made under similar traffic and weather conditions. Consistency of deciding on control action depends on the weather and traffic conditions as well as accuracy in predicting these conditions. Weather parameters (defined in terms of forecast and actual weather and traffic conditions) on different days can be used to categorize days into days with little decision consistency, days with moderate decision consistency and days with high decision consistency. Four years of traffic, weather and ground delay program decisions data at major airports in the United States are used in the analysis. This paper examines performance of different data mining methods in the three regions of decision consistency. Not surprisingly, data mining methods have the best performance in the region of most decision consistency and have the poorest performance in the region of little decision consistency. In applications where data mining methods have differing performance in differing regions, it would be more useful to characterize the region specific performance instead of characterizing performance by a single parameter. Finally, the results show no significant variation in the performance of different data mining methods for this particular problem. The fact that different mining methods show no significant variation also provides further confidence in the results of data mining methods. Work in this abstract discusses initial results. This paper describes the results in terms of both forecast and actual environmental conditions and discusses how prediction errors impact decision consistency.
The RAVEN code is becoming a comprehensive tool to perform probabilistic risk assessment, uncertainty quantification, and verification and validation. The RAVEN code is being developed to support many programs and to provide a set of methodologies and algorithms for advanced analysis. Scientific computer codes can generate enormous amounts of data. To post-process and analyze such data might, in some cases, take longer than the initial software runtime. Data mining algorithms/methods help in recognizing and understanding patterns in the data, and thus discover knowledge in databases. The methodologies used in the dynamic probabilistic risk assessment or in uncertainty and error quantification analysis couple system/physics codes with simulation controller codes, such as RAVEN. RAVEN introduces both deterministic and stochastic elements into the simulation while the system/physics code model the dynamics deterministically. A typical analysis is performed by sampling values of a set of parameter values. A major challenge in using dynamic probabilistic risk assessment or uncertainty and error quantification analysis for a complex system is to analyze the large number of scenarios generated. Data mining techniques are typically used to better organize and understand data, i.e. recognizing patterns in the data. This report focuses on development and implementation of Application Programming Interfaces (APIs) for different data mining algorithms, and the application of these algorithms to different databases.
Text and structural data mining of Web and social media (WSM) provides a novel disease surveillance resource and can identify online communities for targeted public health communications (PHC) to assure wide dissemination of pertinent information. WSM that mention influenza are harvested over a 24-week period, 5-October-2008 to 21-March-2009. Link analysis reveals communities for targeted PHC. Text mining is shown to identify trends in flu posts that correlate to real-world influenza-like-illness patient report data. We also bring to bear a graph-based data mining technique to detect anomalies among flu blogs connected by publisher type, links, and user-tags.