Gotowa bibliografia na temat „Sampling-event datasets”

Dung beetles play key roles in terrestrial ecosystems, contributing to many important ecosystem process and functions, such as nutrient recycling, parasite control and seed dispersal. Due to their tight associations with mammals and their responses to environmental change, they are also frequently used as environmental and biological indicators. Despite their importance, knowledge about dung beetles in Southeast Asia is limited. To address this information gap, we established a databasing project - "Mobilising data on ecologically important insects in Malaysia and Singapore" - funded by the Global Biodiversity Information Facility (GBIF). As part of this project, we compiled two extensive datasets – a sampling-event and occurrence dataset and a taxonomic checklist – for the dung beetles of Sabah, Bornean Malaysia. The sampling-event dataset documents 2,627 unique sampling events and 21,348 dung beetle occurrence records for Sabah. The taxonomic checklist includes 156 confirmed dung beetle species and 36 synonyms, totalling 192 records. These datasets have been made open access through the GBIF portal, which we hope will enhance the understanding of dung beetle taxonomy and their distributions in Southeast Asia.All data presented in this paper comprises of available information pertaining to the dung beetles of Sabah.

A large volume of security events, generally collected by distributed monitoring sensors, overwhelms human analysts at security operations centers and raises an alert fatigue problem. Machine learning is expected to mitigate this problem by automatically distinguishing between true alerts, or attacks, and falsely reported ones. Machine learning models should first be trained on datasets having correct labels, but the labeling process itself requires considerable human resources. In this paper, we present a new selective sampling scheme for efficient data labeling via unsupervised clustering. The new scheme transforms the byte sequence of an event into a fixed-size vector through content-defined chunking and feature hashing. Then, a clustering algorithm is applied to the vectors, and only a few samples from each cluster are selected for manual labeling. The experimental results demonstrate that the new scheme can select only 2% of the data for labeling without degrading the F1-score of the machine learning model. Two datasets, a private dataset from a real security operations center and a public dataset from the Internet for experimental reproducibility, are used.

The present paper describes a sampling-event dataset on species belonging to two families of Diptera (Syrphidae and Asilidae) collected between 2012 and 2019 in two Italian beech forests located in the central Apennines. The reference dataset consists of an annotated checklist and has been published on Zenodo. Syrphidae and Asilidae are two widespread and key ecological groups, including predator, pollinator and saproxylic species. Despite their pivotal role in both natural and man-made ecosystems, these families are still poorly known in terms of local distribution and open-access sampling-event data are rare in Italy.This open-access dataset includes 2,295 specimens for a total of 21 Asilidae and 65 Syrphidae species. Information about the collection (e.g. place, date, methods applied, collector) and the identification (e.g. species name, author, taxon ID) of the species is provided. Given the current biodiversity crisis, the publication of checklists, sampling-event data and datasets on insect communities in open-access repositories is highly recommended, as it represents the opportunity to share biodiversity information amongst different stakeholders. Moreover, such data are also a valuable source of information for nature reserve managers responsible for monitoring the conservation status of protected and endangered species and habitats and for evaluating the effects of conservation actions over time.

Ground beetles (Carabidae, Coleoptera) are one of the most species-rich and well-studied insect families. However, the number of published datasets is disproportionately low against the biodiversity of this group. According to GBIF, only a fifth of the percentage of all published data covers ground beetles. This article describes a sampling-event dataset providing primary data on ground beetles collected in urban and suburban habitats in Kaluga, a typical central Russian city. We surveyed habitats of different land-use types and the extent and intensity of anthropogenic influence: yards, gardens, quarries, small urban woodlands, grasslands and riparian habitats. Carabids were collected by pitfall traps during most of the vegetative season (mostly from late April - early May to at least early October) for 13 seasons between 1994 and 2015. In total, the dataset contains 189 carabid species and 79,091 specimens. The dataset provides information about species composition and abundance, habitat distribution, seasonal and long-term dynamics of carabid beetles in environments of different degrees of urbanisation.This dataset is the first sampling-event dataset about carabids in various urban habitats published through GBIF.

AbstractWith Process discovery algorithms, we discover process models based on event data, captured during the execution of business processes. The process discovery algorithms tend to use the whole event data. When dealing with large event data, it is no longer feasible to use standard hardware in a limited time. A straightforward approach to overcome this problem is to down-size the data utilizing a random sampling method. However, little research has been conducted on selecting the right sample, given the available time and characteristics of event data. This paper systematically evaluates various biased sampling methods and evaluates their performance on different datasets using four different discovery techniques. Our experiments show that it is possible to considerably speed up discovery techniques using biased sampling without losing the resulting process model quality. Furthermore, due to the implicit filtering (removing outliers) obtained by applying the sampling technique, the model quality may even be improved.

The Huai He floodplain in Anhui and Jiangsu Provinces, an important component of the East Asian-Australasian Flyway (EAAF), sustains critical wetland habitats for migratory waterbirds, including four threatened species on the IUCN Red List: critically endangered <i>Aythya baeri</i> (Radde, 1863), endangered <i>Anser cygnoides</i> (Linnaeus, 1758) and vulnerable <i>Melanitta fusca</i> (Linnaeus, 1758) and <i>Aythya ferina</i> (Linnaeus, 1758). Despite its biogeographic significance as a transitional zone between the Yangtze and Yellow River floodplains, this region remains one of China's most understudied and ecologically degraded freshwater systems. Historical pollution events and contemporary anthropogenic pressures – agricultural intensification, hydrological fragmentation and invasive species - have severely compromised wetland integrity. During mid-December 2005 and November to December 2006, standardised surveys employed fixed-radius point counts (158 sites) with the component counting method to enhance accuracy.We present the first comprehensive waterbird dataset for the Anhui and Jiangsu part of the Huai He floodplain, comprising 44 species (32,517 individuals) recorded across 30 wetlands during 2005–2006 surveys. All occurrence data adhere to Darwin Core standards and are accessible via the Global Biodiversity Information Facility, providing spatial-temporal baselines for abundance and distributional data for waterbirds in this region.

DEIMS-SDR (Dynamic Ecological Information Management System - Site and dataset registry, Wohner et al. 2019) is one of the largest repositories of long-term ecological research (LTER) datasets. It provides sophisticated searching tools by metadata elements and identifiers for all the 930 contained datasets, most of them from European sites. Whereas datasets' metadata are highly structured and searchable, datasets themselves have little standardization in terms of content, identifiers or license, making data integration difficult or cumbersome. Adopting the data FAIR guiding principles(Wilkinson et al. 2016) for LTER data would result in better data integration and reutilization to support knowledge discovery and innovation in ecological research. The Global Biodiversity Information Facility (GBIF 2019a). is the largest repository of species distribution data in the world, providing access to more than a billion records from over 43,000 datasets. GBIF is a good example of FAIR principles implementation: GBIF data is highly standardized, using Darwin Core (Wieczorek et al. 2012) for data and ecological metadata language (EML, Fegraus et al. 2005) for metadata, allowing record-level search; and has implemented globally unique and persistent identifiers for datasets and downloads. Relevant in this context is that GBIF has recently introduced a new data format intended for monitoring projects and sampling event protocols (GBIF 2019b). In this presentation, we explore the suitability of GBIF data formats and workflows to serve LTER datasets, and the work it may take to transform typical LTER datasets into these formats. For this exercise, we use some datasets available via the DEIMS platform, corresponding to the same territory, (Sierra Nevada, Spain (e.g. Bonet 2016, Bonet 2018) and transform them into the GBIF's sample-based Event core publish them in the GBIF data network, and then perform an analysis to assess how the standardized datasets work in practice, both among themselves and also with typical "occurrence-based" GBIF datasets. Finally, we discuss our findings and make recommendations for the GBIF and LTER communities.

This paper presents two sampling-event datasets with occurrences of vascular plants, bryophytes and lichens collected in May-June 2021 in southern Ukraine. We aimed to collect high-quality biodiversity data in an understudied region and contribute it to international databases and networks. The study was carried out during the 15th Eurasian Dry Grassland Group (EDGG) Field Workshop in southern Ukraine and the Dark Diversity Network (DarkDivNet) sampling in the Kamianska Sich National Nature Park. By chance, these datasets were collected shortly before the major escalation of the Russian invasion in Ukraine. Surveyed areas in Kherson and Mykolaiv Regions, including established monitoring plots, were severely affected by military actions in 2022. Therefore, collected data are of significant value in the context of biodiversity documentation. The knowledge about the biodiversity of this area will help to assess the environmental impact of the war and plan restoration of the damaged or destroyed habitats. The first preliminary analysis of collected data demonstrates the biodiversity richness and conservation value of studied grassland habitats.We provide sampling-event datasets with 7467 occurrences, which represent 708 taxa (vascular plants, bryophytes and lichens) collected in 275 vegetation relevés. Amongst them, vascular plants are represented by 6665 occurrences (610 taxa), lichens - 420 (46) and bryophytes - 381 (51). Several new species were reported for the first time at the national or regional level. In particular, one vascular plant species (<i>Torilis pseudonodosa</i>) and two lichen species (<i>Cladonia conista</i>, <i>Endocarpon loscosii</i>) were new to Ukraine. One vascular plant (<i>Stipa tirsa</i>), two species of bryophytes (<i>Rhynchostegium megapolitanum</i>, <i>Ptychostomum torquescens</i>) and three species of lichens (<i>Cladonia cervicornis</i>, <i>C. symphycarpa</i>, <i>Involucropyrenium breussi</i>) were recorded for the first time for the Kherson Region. Additionally, these datasets contain occurrences of taxa with narrow distribution, specialists of rare habitat types and, therefore, represented by a low number of occurrences in relevant biodiversity databases and particularly in GBIF. This publication highlights the diversity of natural vegetation and its flora in southern Ukraine and raises conservation concerns.

The Ocean Biogeographic Information System (OBIS) aims to integrate smaller, isolated datasets into a larger, more comprehensive picture of life in our oceans. Therefore, OBIS provides a gateway to many datasets containing information on where and when marine species have been observed. The datasets within OBIS are contributed by a network of hundreds of institutes, projects and individuals, all with the common goal to gain scientific knowledge and to make these data and knowledge easily available to the public. Until recently, OBIS had solely focused on biogeographic data, in the form of presence of marine species in space and time. Data collected for biological studies however often include more than just presence or abundance. Physical and chemical measurements are often taken concomitantly providing insights into the environmental conditions the species live in. Details on the nature of the sampling methods, equipment used and effort can also be of major importance. Based on requirements from the growing OBIS community for data archiving and scientific applications, OBIS completed the OBIS-ENV-DATA project in 2017 to enhance its data standard by accommodating additional data types (De Pooter et al. 2017). The proposed standard allows for the management of sampling methodology, animal tracking and telemetry data, and environmental measurements such as nutrient concentrations, sediment characteristics and other abiotic parameters measured during sampling. The new OBIS data standard builds on the Darwin Core Archive and on practices adopted by the Global Biodiversity Information Facility (GBIF). It consists of an Event Core in combination with an Occurrence Extension and an enhanced MeasurementOrFact Extension Fig. 1. This new structure enables the linkage of measurements or facts - quantitative or qualitative properties - to both sampling events and species occurrences, and includes additional fields for property standardization. The OBIS standard also embraces the use of the new Darwin Core term parentEventID, enabling a sampling event hierarchy. During the follow-up project "OBIS-Event Data", the format will be further fine-tuned during two workshops with two different communities of practice. The first workshop (April 2018) will focus on animal tagging and tracking data, while the second one (October 2018) will tackle macro- and meiobenthos data. The OBIS-Event Data project will also develop the first data products and applications based on the standard and make these tools part of the core OBIS data system output. We believe that the adoption of this new data standard by the international community will be key to improving the effectiveness of the knowledge base and will enhance integration and management of critical data needed to understand ecological and biological processes in the ocean.

Abstract. The Siberian Arctic is warming rapidly, causing permafrost to thaw and altering the biogeochemistry of aquatic environments, with cascading effects on the coastal and shelf ecosystems of the Arctic Ocean. The Lena River, one of the largest Arctic rivers, drains a catchment dominated by permafrost. Baseline discharge biogeochemistry data are necessary to understand present and future changes in land-to-ocean fluxes. Here, we present a high-frequency 4.5-year-long dataset from a sampling program of the Lena River's biogeochemistry, spanning April 2018 to August 2022. The dataset comprises 587 sampling events and measurements of various parameters, including water temperature, electrical conductivity, stable oxygen and hydrogen isotopes, dissolved organic carbon concentration and 14C, colored and fluorescent dissolved organic matter, dissolved inorganic and total nutrients, and dissolved elemental and ion concentrations. Sampling consistency and continuity and data quality were ensured through simple sampling protocols, real-time communication, and collaboration with local and international partners. The data are available as a collection of datasets separated by parameter groups and periods at https://doi.org/10.1594/PANGAEA.913197 (Juhls et al., 2020b). To our knowledge, this dataset provides an unprecedented temporal resolution of an Arctic river's biogeochemistry. This makes it a unique baseline on which future environmental changes, including changes in river hydrology, at temporal scales from precipitation event to seasonal to interannual can be detected.

Dans le cadre de mon travail de thèse, je me suis intéressé aux aérosols atmosphériques et à leur transfert de l’atmosphère vers les surfaces terrestres par les précipitations. La stratégie générale que j’ai suivie repose sur l’observation des dépôts humides sur différentes échelles de temps, interannuelle d’une part et intraévènementielle de l’autre. Elle repose aussi sur leur observation dans des environnements marqués en termes de charge et de composition en aérosols, mais aussi de dynamiques atmosphériques et de précipitations. Le fait de combiner des mesures à la fois sur la composition de l’atmosphère et sur la composition des dépôts humides permet d’identifier la nature des dépôts (intensité, composition, source et provenance) et d’expliquer les phénomènes impliqués dans les dépôts. Cela passe par la documentation complète de différents paramètres (aérosols, dynamique, pluie, dépôt) sur les mêmes périodes de temps, ce qui est néanmoins complexe à mettre en oeuvre. Les deux axes de mon travail portent sur des questions distinctes et complémentaires de l’étude des dépôts humides.Le premier axe s’est porté sur les dépôts humides au Sahel, région semi-aride où le lessivage des poussières minérales de l’atmosphère est un processus clé pour contraindre le bilan atmosphérique en masse de ces composés. Dans cette région marquée par la présence de nombreux systèmes convectifs contrôlant les quantités de précipitations annuelles, la question sur les liens entre dynamiques atmosphériques et dépôts s’est alors posée. La stratégie d’observation long-terme mis en place sur les stations au Sahel dans le cadre du réseau INDAAF, avec une synergie autour de mesures météorologiques, de concentrations et de dépôts d’aérosols, a permis de constituer une base de données très complète. À partir de cette base de données pluriannuelle aux stations de Banizoumbou (Niger) et de Cinzana (Mali) de 2007 et 2015, l’identification de phénomène de cold pools (gouttes froides) à partir de données météorologiques de surface et leur lien avec les retombées de poussières minérales sont discutés. Les ratios de lessivage ont été calculés pour les évènements associés aux cold pools et varient sur plusieurs ordres de grandeur en fonction de l’effet de dilution qui diffère selon les régimes de concentrations atmosphériques en poussière minérale. Les évènements les plus convectifs associés à des concentrations élevées présentent une gamme de valeurs moins dispersée (319 – 766) qui ne dépend pas de la quantité de précipitation.Le second axe s’est focalisé sur l’étude intraévènementielle des dépôts en milieu urbain pour diverses situations de pluie, de concentration et composition en aérosols. Que peut nous apprendre le suivi des dépôts au cours d’un évènement de pluie ? Pour y répondre, j’ai tout d’abord participé au développement d’un collecteur me permettant de collecter les dépôts humides en fractions successives au cours de la pluie. Complétées par un ensemble de mesures colocalisées sur les aérosols et les dynamiques atmosphériques acquises sur le terrain pour 8 cas d’étude, les analyses chimiques des dépôts dissouts et particulaires m’ont permis de discuter à la fois la provenance des aérosols, mais aussi les processus mis en jeu. J’ai pu quantifier la décroissance des concentrations, même de composés traces, dans les dépôts au cours de la pluie. J’ai également pu documenter l’évolution de la solubilité pour les espèces chimiques des dépôts et discuter des poids relatifs des mécanismes de lessivage dans- (rainout) et sous- (washout) le nuage. La variabilité des dépôts observée au cours d’un évènement est au final aussi importante que celle observée entre évènements de pluie<br>In the work conducted for my thesis, I studied atmospheric aerosols and their transfer from the atmosphere to the surface by precipitation. The main strategy I followed is based on the observation of wet deposition on different time scales, interannual on one hand and intra-event on the other. It also relies on their observation in environments marked in terms of aerosol load and composition, but also in terms of atmospheric dynamics and precipitation. Combining measurements on both atmospheric and wet deposition compositions allows to identify the characteristics of the deposition (intensity, composition, source and origin) and to explain the phenomena involved in the deposition. This requires the complete documentation of different parameters (aerosols, dynamics, rainfall, deposition) over the same periods of time, which is nevertheless complex to implement. The two axes of my work deal with distinct and complementary issues in the study of wet deposition.The first focus has been on wet deposition in the Sahel, a semi-arid region where the scavenging of mineral dust from the atmosphere is a key process to constrain the atmospheric mass balance of these compounds. In this region marked by the presence of numerous convective systems controlling annual precipitation amounts, the question of the links between atmospheric dynamics and deposition was addressed. The long-term observation strategy implemented at stations in the Sahel as part of the INDAAF network, with a synergy of meteorological measurements, aerosol concentrations and deposition, has enabled the creation of a very complete database. From this multi-year dataset at Banizoumbou (Niger) and Cinzana (Mali) stations from 2007 and 2015, the identification of cold pools phenomena from surface meteorological data and their link with mineral dust deposition are discussed. Washout ratios have been calculated for cold pool events and vary over several orders of magnitude depending on the dilution effect which differs according to the levels of atmospheric aerosol concentrations. The most convective events associated with high concentrations have a less scattered range of values (319 – 766) that does not depend on the amount of precipitation.The second axis focused on the intra-event study of wet deposition in urban areas for various rainfall situations, aerosol concentration and composition. The question is: what can we learn from the monitoring of deposition during a rain event? To answer this, I first participated in the development of a collector allowing me to collect wet deposition in successive fractions during the rain event. Complemented by a set of co-located measurements on aerosols and atmospheric dynamics acquired in the field for 8 study cases, the chemical analyses of dissolved and particulate deposition allowed me to discuss both the origin of the aerosols and processes involved. I was able to quantify the decay of concentrations, even of trace compounds, in the deposits during rainfall. I was also able to document the evolution of solubility for chemical species in the deposition and discuss the relative contribution of the rainout and washout mechanisms. The variability of deposition observed during an event is actually as significant as that observed between rain events

Digital is constantly evolving with the appearance of connected objects and on top of the popularization today of artificial intelligence. One of the direct inductions remains the excessive proliferation of various kinds of attacks in computer systems. Hackers exploit these vulnerabilities to break in and attack systems with increasingly complex attacks. The consequences of intrusions are destructive and ruinous for businesses and organizations such as electronic ransom ware, data alteration and loss, financial and brand image loss.It is important for those involved in computer systems to equip any computer centre with adequate tools to prevent malicious individuals from accessing the systems. To remedy these setbacks, several IT tools are developed including IDS intrusion detection systems. IDS intrusion detection systems are devices designed to monitor a computer system, give alerts and trigger real-time counterattacks in the event of attacks. These intelligent systems use several detection approaches and various algorithms. The performance of the IDS is increased when the features dimensionality are reduced significantly.This study proposed feature dimensionality reduction techniques such as Principal Component Analysis (PCA) and Auto-Encoder (AE). The output from the reduced dimensional features are used to build machine Learning algorithms. The performance results is evaluated on the CSECICIDS2018 datasets. The proposed public intrusion data sets suffer from the Imbalance class. In order to handle this issue, we propose hybrid sampling technique by combining Over and undersampling technique.The performance results from the reduced features in terms of true positive, False positve, recall, precision, F-Measure, ROC Area, PRC Area show the better performance. In addition, the obtained results are compared with deep convolutional approach.

ABSTRACT Distributed acoustic sensing (DAS) can provide high-resolution measurements owing to the closely spaced sensing channels over a long distance and high sampling rate – attributes that are beneficial to seismic monitoring and analysis. However, DAS arrays generally suffer from a lower signal-to-noise ratio (SNR). Additionally, dense spatial and temporal measurements can result in extremely large data volumes that hinder efficient data storage, transmission, and processing. Automatically and accurately detecting seismic events in continuous DAS data is challenging. In this study, we present a microseismic event detection workflow that is based on coherency analysis of the waveforms and accommodates the high-resolution characteristics of DAS data. DAS data is transformed into an apparent velocity-time (v-t) domain by slant-stacking the data along different apparent velocities, i.e., slopes of the waveform, at each time sample. The coherency of the slant-stacked waveform is measured by the semblance coefficient. A coherent signal is identified if the semblance coefficient is larger than a threshold that is adaptively calculated. Then, the density-based spatial clustering (DBSCAN) algorithm is applied to group the coherent detections into different clusters. Finally, the microseismic events are detected by filtering out the clusters of coherent noise. The algorithm generates low-volume time-windowed data containing microseismic events that can be efficiently transferred and processed to estimate event location, magnitude, and source mechanism. The performance of the developed algorithm is tested using two field datasets from the Utah FORGE project. The results demonstrate the potential of the algorithm to achieve automatic real-time microseismic event detection for downhole DAS systems. INTRODUCTION Microseismic monitoring is a critical element for a variety of subsurface operations such as hydraulic fracturing in unconventional reservoirs, water circulation in Enhanced Geothermal Systems (EGS), as well as carbon injection and storage in geological reservoirs. In hydraulic fracturing treatments, microseismicity is important for mapping and characterizing created fracture networks. In fluid injection and storage operations, potential seismic hazard may occur due to the large volume of injected fluid. Seismic monitoring, being also a regulatory requirement, offers an early warning tool to avoid and mitigate seismic hazard, also known as traffic light monitoring system. Seismic event detection is the first step for subsequent processing encompassing event location, magnitude estimation, and source mechanism characterization. Recently, distributed acoustic sensing (DAS) has been increasingly used for microseismic monitoring. DAS makes use of Rayleigh scattering of laser pulses in fiber-optic cables to measure strain rate along a fiber (Hartog, 2017). One significant advantage of DAS is the high spatial-temporal resolution and large coverage. The fiber can be kilometers long and the channel spacing is in the order of meters, such that thousands of measurements can be obtained at each time sample. The sampling rate of DAS can be thousands of Hertz (Hz). Compared with the traditional geophone or accelerometer-type sensors, where there are only tens of sensing points with large inter-sensor spacing, DAS can provide a more complete picture of microseismic waveforms propagating in time and space. However, the dense spatial and temporal measurements generate extremely large data volumes. Furthermore, the signal-to-noise ratio of DAS data is generally lower than that of the geophone-acquired data. These challenges require the development of new methods for automated real-time microseismic monitoring and analysis.

Abstract Early detection and characterization of anomalous events during drilling operations are critical to avoid costly downtime and prevent hazardous events, such as a stuck pipe or a well control event. A key aspect of real-time drilling data analysis is the capability to make precise predictions of specific drilling parameters based on past time series information. The ideal models should be able to deal with multivariate time series and perform multi-step predictions. The recurrent neural network with a long short-term memory (LSTM) architecture is capable of the task, however, given that drilling is a long process with high data sampling frequency, LSTMs may face challenges with ultra-long-term memory. The transformer-based deep learning model has demonstrated its superior ability in natural language processing and time series analysis. The self-attention mechanism enables it to capture extremely long-term memory. In this paper, transformer-based deep learning models have been developed and applied to real-time drilling data prediction. It comprises an encoder and decoder module, along with a multi-head attention module. The model takes in multivariate real-time drilling data as input and predicts a univariate parameter in advance for multiple time steps. The proposed model is applied to the Volve field data to predict real-time drilling parameters such as mud pit volume, surface torque, and standpipe pressure. The predicted results are observed and evaluated. The predictions of the proposed models are in good agreement with the ground truth data. Four Transformer-based predictive models demonstrate their applicability to forecast real-time drilling data of different lengths. Transformer models utilizing non-stationary attention exhibit superior prediction accuracy in the context of drilling data prediction. This study provides guidance on how to implement and apply transformer-based deep learning models applied to drilling data analysis tasks, with a specific focus on anomaly detection. When trained on dysfunction-free datasets, the proposed model can predict real-time drilling data with high precision, whereas when a downhole anomaly starts to build, the significant error in the prediction can be used as an alarm indicator. The model can consider extremely long-term memory and serve as the alternative algorithm to LSTM. Furthermore, this model can be extended to a wide range of sequence data prediction problems in the petroleum engineering discipline.

We develop a method for the evaluation of extreme event statistics associated with nonlinear dynamical systems, using a very small number of samples. From an initial dataset of design points, we formulate a sequential strategy that provides the ‘next-best’ data point (set of parameters) that when evaluated results in improved estimates of the probability density function (pdf) for a scalar quantity of interest. The approach utilizes Gaussian process regression to perform Bayesian inference on the parameter-to-observation map describing the quantity of interest. We then approximate the desired pdf along with uncertainty bounds utilizing the posterior distribution of the inferred map. The ‘next-best’ design point is sequentially determined through an optimization procedure that selects the point in parameter space that maximally reduces uncertainty between the estimated bounds of the pdf prediction. Since the optimization process utilizes only information from the inferred map it has minimal computational cost. Moreover, the special form of the criterion emphasizes the tails of the pdf. The method is applied to estimate the extreme event statistics for a very high-dimensional system with millions degrees of freedom: an offshore platform subjected to three-dimensional irregular waves. It is demonstrated that the developed approach can accurately determine the extreme event statistics using orders of magnitude smaller number of samples compared with traditional approaches.

Abstract Traceability of food products to their sources is critical for quick responses to food emergencies. However, having the complete and consistent information needed to quickly investigate sources and identify affected material has proven difficult. Food trace-ability is challenging for a variety of reasons including diversity and heterogenicity of participants, complexity of the supply chain and its processes and lack of a common understanding of steps in a supply chain, and incompleteness of data, and unwillingness of actors to expose information of their internal operations. The objective of this work is to address the traceability challenge by developing a formal ontology that can provide a shared and common understanding of the traceability model across all stakeholders in a food supply chain. In previous research, an ontological approach was adopted to address the traceability problem in the context of a use case related to harvest to on-farm storage activities. This work extends the Supply Chain Traceability ontology by introducing additional critical tracking events including transform event, sampling event, observation event, custody change event, and ownership change event in the context of a scenario involving shipments of commodity grain from a primary grain elevator to a processor of grain such as a feed manufacturer. A knowledge graph is generated based on a simulated dataset and the ontology is validated through query, reasoning, and visualization conducted in the RDFox environment.

&lt;div class="section abstract"&gt;&lt;div class="htmlview paragraph"&gt;Automated driving systems (ADS) are designed toward safely navigating the roadway environment, which also includes consideration of potential conflict with other road users. Of particular concern is understanding the cumulative risk associated with vulnerable road users (VRUs) conflicts and collisions. VRUs represent a population of road users that have limited protection compared to vehicle occupants. These severity distributions are particularly useful in evaluating ADS real-world performance with respect to the existing fleet of vehicles. The objective of this study was to present event severity distributions associated with vehicle-cyclist collisions within an urban naturalistic driving environment by leveraging data from third-party vehicles instrumented with forward-facing cameras and a sensor suite (accelerometer sampling at 20 Hz and GPS [variable sampling frequency]). From over 66 million miles of driving, 30 collision events were identified. A global optimization routine was used on the accelerometer and GPS data to correct for sensor orientation and asynchronicity in data sampling. For each event, two key video frames were identified: the frame associated with impact and a frame associated with key vehicle kinematics (e.g. vehicle start/stop). These key frames were then mapped to the accelerometer and GPS data to determine vehicle speed at impact. For the events included in this dataset, impact speeds ranged from approximately 3.2 kph (2 mph) to 53.1 kph (33 mph). In 82% of events, the front of the vehicle struck the cyclist. Existing cyclist injury risk curves were then used to calculate the level of risk associated with the reconstructed impacts, and the probability of AIS3+ injury risk was observed to vary from minimal risk to approximately 30%. These data highlight the wide range of impact speeds and injury risk that may occur during vehicle-cyclist collisions.&lt;/div&gt;&lt;/div&gt;