Large Data Streams Research Articles

Архитектура для непрерывного извлечения знаний из социальных сетей

Nowadays social networks and media play significant role in daily life. All our life in the real world is recorded in the digital space as well. Scientists have enormous potential in researching issues such as social influence ontop news and top news influence on society. Its impact on daily life spans such diverse areas as digital marketing, publicopinion analysis, political monitoring and disaster notification. Any task of processing such a large data stream needs acoherent architecture that will fit the analyzed resource. In the presented work, we set ourselves the task of creating ahighly loaded, fault-tolerant, scalable system for extracting and processing data from various social networks and analyzing data in real time. The solution is architecture in the form of a set of modules. Modules have their own characteristics depending on the work performed, from collecting textual data to direct processing and extraction of knowledge.

A Systematic Review of Density Grid-Based Clustering for Data Streams

Various applications, such as electronic business, satellite remote sensing, intrusion discovery, and network traffic monitoring, generate large unbounded data stream sequences at a rapid pace. The clustering of data streams has attracted considerable interest due to the increasing usage of evolving data streams. In particular, evolving data streams affect clustering because they introduce numerous challenges, such as time and memory limits and one-pass clustering. Furthermore, researchers need to be able to determine arbitrarily shaped clusters present in evolving data streams from applications. Due to these characteristics, conventional density grid-based clustering techniques cannot be used. Moreover, the existing density grid-based clustering algorithms have low cluster quality for clustering evolving data streams. This study conducted a systematic literature review (SLR) and noted numerous research-related issues encountered in solving the aforementioned problems. We summarized numerous grid-based clustering algorithms that have been used and determined their distinctive and limited features. We also observed how these algorithms address the challenges affecting the clustering of evolving data streams and studied their advantages and disadvantages. SLR was based on 104 articles published between 2010 and 2021. Numerous challenges remain for grid-based clustering algorithms, particularly in terms of time-limited and high-dimensional data handling. Last, our findings indicated a variety of active studies on density grid-based clustering.

Enhanced-KNN (M-KNN) based outlier detection and sensor data aggregation for large data streams in the IoT-cloud

Enhanced-KNN (M-KNN) based outlier detection and sensor data aggregation for large data streams in the IoT-cloud

ИНТЕЛЛЕКТУАЛЬНАЯ ОБРАБОТКА БОЛЬШИХ ДАННЫХ В ЗАДАЧАХ МАЛОГО БИЗНЕСА НА ОСНОВЕ БАЙЕСОВСКИХ ИНТЕЛЛЕКТУАЛЬНЫХ ТЕХНОЛОГИЙ

The article is devoted to solving the urgent problem of creating intelligent big data processing systems for small businesses. The specificity of information flows of small businesses, in particular in the service sector, is shown.The main tasks of increasing the efficiency of small businesses based on digitalization are identified. The methodology of the regularizing Bayesian approach (RBP), Bayesian intelligent measurement technologies (BII) and a digital platform for managing small businesses and creating monitoring and decision support systems are proposed. Based on this platform of the RBP methodology, a working sample of a complex for service enterprises – fitness centers has been developed. The unique properties of an intelligent complex are the possibilities of integrating different types of information, intelligent processing of large data streams in the form of numerical and linguistic information under conditions of considerable uncertainty In the software package for the service sector, various situations can be simulated when working with clients, as well as digital images of clients can be created to personalize work with each of them for the organization of effective and profitable work of small business companies.

Explaining smartphone-based acoustic data in bipolar disorder: Semi-supervised fuzzy clustering and relative linguistic summaries

Smartphones enable to collect large data streams about phone calls that, once combined with Computational Intelligence techniques, bring great potential for improving the monitoring of patients with mental illnesses. However, the acoustic data streams recorded in uncontrolled environments are dynamically changing due to various sources of uncertainty. In addition, such acoustic data are usually difficult to interpret by psychiatrists. Within this study, we propose an approach based on Linguistic Summaries with Fuzzy Clustering (LS-FC) aiming at the development of human-consistent and easily interpretable summaries about relations between acoustic data and mental state of a patient affected by Bipolar Disorder, e.g., Most calls in the state of hypomania have low loudness compared to the state of euthymia [T = 1]. To capture the dynamics of acoustic data streams, we apply a dynamic incremental semi-supervised fuzzy clustering that synthesizes data into clusters. These clusters are represented by prototypes which are used for the construction of the membership functions describing linguistic terms e.g., low loudness, and then, linguistic summaries. The main contribution of this paper is the incorporation of information about clusters’ prototypes in the generation of linguistic summaries. The primary goal of this research is explainability. The semi-supervised learning algorithm is used mainly for deriving clusters and building improved linguistic summaries. Numerical results indicate that linguistic summaries provide intuitive and clear information about voice features in a patient’s affective state and they are consistent with clinical observation. In particular, during most calls in hypomania/mania both the quality of the patient’s voice and the dynamics of change in the spectrum signal reflected in spectral flux are low compared to euthymia. The proposed approach enables to summarize large data streams into meaningful descriptions that, although relatively simple, offer information granules that are very intuitive for clinicians and are promising to support the smartphone-based monitoring of bipolar disorder patients to inform about the potential change of mental state.

Tardiness Bounds for Sporadic Gang Tasks Under Preemptive Global EDF Scheduling

Following the trend of increasing autonomy in cyber-physical systems, parallel embedded architectures have enabled devices to better handle the large streams of data and intensive computation required by such autonomous systems. However, while the explosion of highly-parallel platforms has seen a proportional growth in the number of applications/devices that utilize these platforms, the embedded systems community's understanding of how to build time-predictable, safety-critical systems with parallel platforms has not kept pace. As a well-motivated but challenging parallel scheduling model, gang scheduling requires all parallel threads of each parallel task to simultaneously execute in unison, which is in contrast to traditional, multi-threaded parallel scheduling, where a parallel task may spawn multiple threads, and each thread will be scheduled independently of other threads of the same task. While increasing research efforts on hard real-time (HRT) gang scheduling have recently been seen, the problem of gang scheduling in the context of soft real-time (SRT) systems, where provably bounded deadline tardiness can be tolerated, has hardly been studied yet. In this article, we derive and prove the first tardiness bounds for sporadic gang task systems under preemptive GEDF scheduling. A total utilization bound for SRT-schedulability is required for ensuring such tardiness bounds but it is shown to be tight with respect to the platform capacity and maximum parallelism-induced idleness. Furthermore, we also empirically evaluate the effects of different degrees of task parallelism upon the SRT-schedulability.

Efficient strategies for incremental mining of frequent closed itemsets over data streams

Mining frequent closed itemsets over data streams is an important data mining problem. Mining data streams is more challenging than mining static data because of the nature of data streams, including high arrival rate, massive volume of incoming data, and concept drift. The existing algorithms for mining frequent closed itemsets over data streams suffer from scalability and efficiency bottlenecks. This paper proposes a novel algorithm for mining frequent closed itemsets over data streams both for the sliding window model and for the landmark model. An indexed prefix closed itemset tree is proposed for compressing all closed itemsets and for quick searching of closed itemsets, and novel search strategies are proposed to prune the search space in updating the set of closed itemsets. The proposed algorithm outperforms the state-of-the-art intersection-based algorithms, CICLAD, ConPatSet, and CloStream, by several times to 2 orders of magnitude in efficiency, and also outperforms the state-of-the-art pattern enumeration algorithm, Moment, by up to 2 orders of magnitude over data streams with large windows and sparse data streams. The proposed algorithm is also superior in scalability.

COVID-19 Telehealth Impact Study

Importance: This three-part study characterizes the widespread implementation of telehealth during the first year of the COVID-19 pandemic, giving us insight into the role of telehealth as we enter a stage of “new normal” healthcare delivery in the U.S.
 Objective: The COVID-19 Telehealth Impact Study was designed to describe the natural experiment of telehealth adoption during the pandemic. Using a large claims data stream and surveys of providers and patients, we studied telehealth in all 50 states to inform healthcare leaders. 
 Design, Setting, Participants: In March 2020, the MITRE Corporation and Mayo Clinic founded the COVID-19 Healthcare Coalition (C19HCC), to respond to the pandemic. We report trends using a dataset of over 2 billion healthcare claims covering over 50% of private insurance activity in the U.S. (January 2019-December 2020), along with key elements from our provider survey (July-August 2020) and patient survey (November 2020 - February 2021).
 Main Outcomes and Measures: There was rapid and widespread adoption of telehealth in Spring 2020 with over 12 million telehealth claims in April 2020, accounting for 49.4% of total health care claims. Providers and patients expressed high levels of satisfaction with telehealth. 75% of providers indicated that telehealth enabled them to provide quality care. 84% of patients agreed that quality of their telehealth visit was good.
 Results: Peak levels of telehealth use varied widely among states ranging from 74.9% in Massachusetts to 25.4% in Mississippi. Every clinical discipline saw a steep rise with the largest claims volume in behavioral health. Provision of care by out-of-state provider was common at 6.5% (October-December 2020). Providers reported multiple modalities of telehealth care delivery. 74% of patients indicated they will use telehealth services in the future.
 Conclusions and Relevance: Innovation shown by providers and patients during this period of rapid telehealth expansion constitutes a great natural experiment in care delivery with evidence supporting widespread clinical adoption and satisfaction on the part of both patients and clinicians. The authors encourage continued broad access to telehealth over the next 12 months to allow telehealth best practices to emerge, creating a more effective and resilient system of care delivery.

Scalable teacher forcing network for semi-supervised large scale data streams

The large-scale data stream problem refers to high-speed information flow which cannot be processed in scalable manner under a traditional computing platform. This problem also imposes expensive labelling cost making the deployment of fully supervised algorithms unfeasible. On the other hand, the problem of semi-supervised large-scale data streams is little explored in the literature because most works are designed in the traditional single-node computing environments while also being fully supervised approaches. This paper offers Weakly Supervised Scalable Teacher Forcing Network (WeScatterNet) to cope with the scarcity of labelled samples and the large-scale data streams simultaneously. WeScatterNet is crafted under distributed computing platform of Apache Spark with a data-free model fusion strategy for model compression after parallel computing stage. It features an open network structure to address the global and local drift problems while integrating a data augmentation, annotation and auto-correction (DA3) method for handling partially labelled data streams. The performance of WeScatterNet is numerically evaluated in the six large-scale data stream problems with only 25% label proportions. It shows highly competitive performance even if compared with fully supervised learners with 100% label proportions.

Novel Applications of Technology for Advancing Tidal Marsh Ecology

Over the last 20 years, innovations have led to the development of exciting new technologies and novel applications of established technologies, collectively increasing the scale, scope, and quality of research possible in tidal marsh systems. Thus, ecological research on marshes is being revolutionized, in the same way as ecological research more generally, by the availability of new tools and analytical techniques. This perspective highlights current and potential applications of novel research technologies for marsh ecology. These are summarized under several themes: (1.) imagery — sophisticated imaging sensors mounted on satellites, drones, and underwater vehicles; (2.) animal tracking — acoustic telemetry, passive integrated transponder (PIT) tags, and satellite tracking, and (3.) biotracers — investigation of energy pathways and food web structure using chemical tracers such as compound-specific stable isotopes, isotope addition experiments, contaminant analysis, and eDNA. While the adoption of these technological advances has greatly enhanced our ability to examine contemporary questions in tidal marsh ecology, these applications also create significant challenges with the accessibility, processing, and synthesis of the large amounts of data generated. Implementation of open science practices has allowed for greater access to data. Newly available machine learning algorithms have been widely applied to resolve the challenge of detecting patterns in massive environmental datasets. The potential integration on digital platforms of multiple, large data streams measuring physical and biological components of tidal marsh ecosystems is an opportunity to advance science support for management responses needed in a rapidly changing coastal landscape.

Multiblock variable influence on orthogonal projections (MB-VIOP) for enhanced interpretation of total, global, local and unique variations in OnPLS models

BackgroundFor multivariate data analysis involving only two input matrices (e.g., X and Y), the previously published methods for variable influence on projection (e.g., VIPOPLS or VIPO2PLS) are widely used for variable selection purposes, including (i) variable importance assessment, (ii) dimensionality reduction of big data and (iii) interpretation enhancement of PLS, OPLS and O2PLS models. For multiblock analysis, the OnPLS models find relationships among multiple data matrices (more than two blocks) by calculating latent variables; however, a method for improving the interpretation of these latent variables (model components) by assessing the importance of the input variables was not available up to now.ResultsA method for variable selection in multiblock analysis, called multiblock variable influence on orthogonal projections (MB-VIOP) is explained in this paper. MB-VIOP is a model based variable selection method that uses the data matrices, the scores and the normalized loadings of an OnPLS model in order to sort the input variables of more than two data matrices according to their importance for both simplification and interpretation of the total multiblock model, and also of the unique, local and global model components separately. MB-VIOP has been tested using three datasets: a synthetic four-block dataset, a real three-block omics dataset related to plant sciences, and a real six-block dataset related to the food industry.ConclusionsWe provide evidence for the usefulness and reliability of MB-VIOP by means of three examples (one synthetic and two real-world cases). MB-VIOP assesses in a trustable and efficient way the importance of both isolated and ranges of variables in any type of data. MB-VIOP connects the input variables of different data matrices according to their relevance for the interpretation of each latent variable, yielding enhanced interpretability for each OnPLS model component. Besides, MB-VIOP can deal with strong overlapping of types of variation, as well as with many data blocks with very different dimensionality. The ability of MB-VIOP for generating dimensionality reduced models with high interpretability makes this method ideal for big data mining, multi-omics data integration and any study that requires exploration and interpretation of large streams of data.

A proposal to leverage high-quality veterinary diagnostic laboratory large data streams for animal health, public health, and One Health.

Test data generated by ~60 accredited member laboratories of the American Association of Veterinary Laboratory Diagnosticians (AAVLD) is of exceptional quality. These data are captured by 1 of 13 laboratory information management systems (LIMSs) developed specifically for veterinary diagnostic laboratories (VDLs). Beginning ~2000, the National Animal Health Laboratory Network (NAHLN) developed an electronic messaging system for LIMS to automatically send standardized data streams for 14 select agents to a national repository. This messaging enables the U.S. Department of Agriculture to track and respond to high-consequence animal disease outbreaks such as highly pathogenic avian influenza. Because of the lack of standardized data collection in the LIMSs used at VDLs, there is, to date, no means of summarizing VDL large data streams for multi-state and national animal health studies or for providing near-real-time tracking for hundreds of other important animal diseases in the United States that are detected routinely by VDLs. Further, VDLs are the only state and federal resources that can provide early detection and identification of endemic and emerging zoonotic diseases. Zoonotic diseases are estimated to be responsible for 2.5 billion cases of human illness and 2.7 million deaths worldwide every year. The economic and health impact of the SARS-CoV-2 pandemic is self-evident. We review here the history and progress of data management in VDLs and discuss ways of seizing unexplored opportunities to advance data leveraging to better serve animal health, public health, and One Health.

Sen2Grass: A Cloud-Based Solution to Generate Field-Specific Grassland Information Derived from Sentinel-2 Imagery

Grasslands are important for their ecological values and for agricultural activities such as livestock production worldwide. Efficient grassland management is vital to these values and activities, and remote sensing technologies are increasingly being used to characterize the spatiotemporal variation of grasslands to support those management practices. For this study, Sentinel-2 satellite imagery was used as an input to develop an open-source and automated monitoring system (Sen2Grass) to gain field-specific grassland information on the national and regional level for any given time range as of January 2016. This system was implemented in a cloud-computing platform (StellaSpark Nexus) designed to process large geospatial data streams from a variety of sources and was tested for a number of parcels from the Haus Riswick experimental farm in Germany. Despite outliers due to fluctuating weather conditions, vegetation index time series suggested four distinct growing cycles per growing season. Established relationships between vegetation indices and grassland yield showed poor to moderate positive trends, implying that vegetation indices could be a potential predictor for grassland biomass and chlorophyll content. However, the inclusion of larger and additional datasets such as Sentinel-1 imagery could be beneficial to developing more robust prediction models and for automatic detection of mowing events for grasslands.

Reliability of Automotive Multidomain Controllers: Advancements in electronics cooling technologies

The contemporary automotive market is undergoing rapid changes that will inevitably revolutionize the road transportation sector in the years to come. Alongside electric vehicles and sophisticated vehicular communication systems, autonomous driving (AD) is emerging as a game-changing technology of the near future. To enable this cutting-edge functionality, passenger vehicles are equipped with various data acquisition and processing systems that collect information from vehicle surroundings. Onboard sensors, such as cameras, lidar, and radar, produce large data streams that are integrated and interpreted by processing units and subsequently used to generate signals to control car behavior. To ensure an acceptable level of safety in on-road operating conditions, these operations are performed in (near) real time. Additionally, a typical modern vehicle includes a combination of communication, driver assistance, active safety, and sophisticated infotainment devices, each of which is often a state-of-the-art electronic system [1].

Hybrid Blockchain Architecture for Cloud Manufacturing-as-a-service (CMaaS) Platforms with Improved Data Storage and Transaction Efficiency

Blockchain based decentralized Cloud Manufacturing-as-a-Service (CMaaS) platforms enable customers to gain access to a large capacity of manufacturing nodes over cryptographically secure networks. In recent times, the Ethereum network has emerged as a popular blockchain framework for providing provenance and traceability of proprietary manufacturing data in decentralized CMaaS. However, the Ethereum ecosystem was only designed to store and transmit low volume financial transaction data and little has been done to make it an efficient repository of large manufacturing data streams in CMaaS systems. In this paper, the authors build on their previous work and report the design, implementation, and validation of middleware software architectures that allow Ethereum based distributed CMaaS platforms to harness the benefits of the secure asset models of the Ethereum ecosystem and the immutable big data storage capabilities of the decentralized BigchainDB database platform. A novel hybrid blockchain architecture enabled by efficient communication protocols and blockchain oracles is proposed. This architecture allows the transfer and immutable storage of large manufacturing data streams onto global BigchainDB nodes allowing data rich manufacturing transactions to bypass the transaction fees of the Ethereum ecosystem. Additionally, a machine learning based time series inference model is proposed which enables the forecast of Ethereum gas price into the future. This allows the CMaaS platform smart contracts to judiciously assign gas price limits and hence save on transactions ensuing from transfer or creation of assets. The outcomes of this research show that the designed hybrid architecture can lead to the reduction of significant number of computational steps and hence transaction fees on Ethereum by offloading large volume data onto BigchainDB nodes. A Random Forest regressor based time series inference model has been shown to exhibit superior performance in the prediction of Ethereum gas price, that allows the CMaaS to avoid executing transactions in periods of high gas prices within the Ethereum ecosystem.

Comparison of learning analytics and educational data mining: A topic modeling approach

Educational data mining and learning analytics, although experiencing an upsurge in exploration and use, continue to elude precise definition; the two terms are often interchangeably used. This could be owing to the fact that the two fields exhibit common thematic elements. One avenue to provide clarity, uniformity, and consistency around the two fields, is to identify similarities and differences in topics between the two evolving fields. This paper conducted a topic modeling analysis of articles related to educational data mining and learning analytics to reveal thematic features of the two fields. Specifically, we employed structural topic modeling to identify the topics of the two fields from the abstracts. We apply structural topic modeling on N=192 articles for educational data mining and N=489 articles for learning analytics. We infer five-topic models for both educational data mining and learning analytics. We find that while there appears to be disciplinary differences in terms of research focus, there is little support for a clear distinction between the two disciplines, beyond their different lineage. The trend points to a convergence within the field of educational research on the applications of advanced statistical learning techniques to extract actionable insights from large data streams for optimizing teaching and learning. Both fields have converged on an increasing focus on student behaviors over the last five years.

Rich feature hierarchies for accurate object detection and semantic segmentation

Formulation of the problem. Over the past few years, there has been little progress in object detection techniques. The most efficient are complex computational ensemble methods, which usually combine several low-level image properties with high-level properties. However, every day interest in artificial intelligence is growing, and the idea of using neural networks on board a spacecraft, with the possibility of making decisions and issuing one-time commands, is very promising, since it makes it possible to analyze a large data stream in real time without resorting to ground station, thereby not losing information when transmitting a packet. The purpose of the work is to conduct research on the possibility of effective use of modern models of neural networks, to develop a methodology for their use in the problem of object detection and analysis of the element base for hardware implementation with the possibility of using convolutional neural networks for thermovideotelemetry on board a spacecraft. Results of work. An approach has been formulated that combines two key ideas: 1) application of high-throughput convolutional neural networks for downward processing of image regions in order to localize and segment objects; 2) preliminary training for the auxiliary task, followed by fine tuning of the domain, which gives a significant increase in performance in the case when the training data is insufficient. The analysis of the element base for the possibility of hardware implementation of neural networks on board a spacecraft using electrical radio products of domestic and foreign production is carried out. Practical significance. The efficiency of preliminary network training for an auxiliary task is shown, followed by fine tuning of the subject area. A technique is described that makes it possible to increase the average accuracy of detecting objects in an image by more than 30%. The analysis of the existing element base, the possibility of hardware implementation of neural networks on board the spacecraft using electrical radio products of domestic and foreign production, as well as the criteria for selecting key elements.

An adaptive nonparametric exponentially weighted moving average control chart with dynamic sampling intervals

AbstractNonparametric statistical control charts have been widely studied over the past decade. The variable sampling intervals (VSIs) methods have been found to exhibit superior performance compared to the traditional fixed sampling policy in this area. In previous VSI studies, the sampling interval function d(·) realized only two distinct values. Limiting the selection of the sampling interval to two fixed values may not be reasonable, especially when the monitoring statistic value is at the edge or near the middle of these two intervals so that the advantages of VSI methods cannot be fully reflected. Therefore, we propose a nonparametric exponentially weighed moving average control chart with a dynamic sampling interval in this article, in which the dynamic sampling interval is an extension of two fixed sampling intervals. The dynamic sampling interval can be continuous valued and is determined by an indicator of the magnitude of the unknown shift. This chart can efficiently detect various magnitudes of shifts and performs robustly for different distributions according to simulation studies. The small computational cost indicates that the method is applicable for monitoring large data streams. Data from an aluminum electrolytic capacitor manufacturing process are used to illustrate the implementation of this chart.

Knowledge Discovery Using Topological Analysis for Building Sensor Data.

Distributed sensor networks are at the heart of smart buildings, providing greater detail and valuable insights into their energy consumption patterns. The problem is particularly complex for older buildings retrofitted with Building Energy Management Systems (BEMS) where extracting useful knowledge from large sensor data streams without full understanding of the underlying system variables is challenging. This paper presents an application of Q-Analysis, a computationally simple topological approach for summarizing large sensor data sets and revealing useful relationships between different variables. Q-Analysis can be used to extract novel structural features called Q-vectors. The Q-vector magnitude visualizations are shown to be very effective in providing insights on macro behaviors, i.e., building floor behaviors in the present case, which are not evident from the use of unsupervised learning algorithms applied on individual terminal units. It has been shown that the building floors exhibited distinct behaviors that are dependent on the set-point distribution, but independent of the time and season of the year.

Nuclear data uncertainty propagation and modeling uncertainty impact evaluation in neutronics core simulation

Uncertainty analysis is a critical requirement in reactor simulation as it is used to quantify the reliability of best-estimate calculation. A comprehensive uncertainty analysis should characterize all sources of uncertainties in a computationally-feasible and scientifically-defendable manner. This manuscript employs a well-established reduced order modeling (ROM) based uncertainty quantification methodology to propagate uncertainties throughout neutronic calculations. ROM relies on recent advances in randomized data mining techniques applied to large data streams. In our proposed implementation, the nuclear data uncertainties are first propagated from multi-group level through lattice physics calculation to generate few-group parameter uncertainties, described using a vector of mean values and a covariance matrix. Employing an ROM-based compression of the covariance matrix, the few-group uncertainties are then propagated through downstream core simulation in a computationally efficient manner. This straightforward approach, albeit efficient as compared to brute force forward and/or adjoint-based methods, often employs a number of assumptions that have been unquestioned in the literature of neutronic uncertainty analysis. This manuscript argues that these assumptions could introduce another source of uncertainty referred to as modeling uncertainties, whose magnitude needs to be quantified in tandem with nuclear data uncertainties. Thus, our primary goal is to explore the interactions between these two uncertainty sources in order to assess whether modeling uncertainties have an impact on parameter uncertainties. To explore this endeavor, the impact of a number of modeling assumptions on core attributes uncertainties is quantified. The study employs a CANDU reactor model, with Serpent and NEWT as lattice physics solvers and NESTLE-C as core simulator. The modeling assumptions investigated include those related with the uncertainty propagation method employed, e.g., deterministic vs. stochastic, the few-group energy structure employed to represent the cross-sections, the resonance treatment in lattice physics calculation, the reference values for the cross-section, and the number of samples employed to render ROM compression. Results indicate that some of the modeling assumptions could have a non-negligible impact on the core responses propagated uncertainties, highlighting the need for a more comprehensive approach to combine parameter and modeling uncertainties.