The impact of unknown extra parameters on scatter matrix estimation and detection performance in complex t-distributed data

<p>Service-oriented architecture (SOA) encourages the creation of modular applications involving Web services as the reusable components. Data-intensive Web services have emerged to manipulate and deal with the massive data emerged from technological advances and their various applications. Distributed Data-intensive Web Service Composition (DWSC) is a core of SOA, which includes the selection of data-intensive Web services from diverse locations on the network and composes them to accomplish a complicated task. As a fundamental challenge for service developers, service compositions must fulfil functional requirements and optimise Quality of Service (QoS), simultaneously. The QoS of a distributed DWSC is not only impacted by the QoS of component services and how the compositions are generated, but also by the locations of services and data transformation between services. However, existing works often neglect the impact of locations and data on service composition. The distributed DWSC has not been sufficiently studied in the literature. In this thesis, we first define the single-objective distributed DWSC that includes communication (e.g. bandwidth), Web service (execution time) and data (data cost) attributes. To this aim, we consider bandwidth information of communication links obtained using the location information of services. Based on the problem formulation, we then address the distributed DWSC problem by developing EC-based approaches. Those EC-based approaches are designed to incorporate domain-knowledge for effectively solving the distributed DWSC problem. Afterwards, we study the multi-objective distributed DWSC to satisfy different QoS requirements. In particular, the QoS-constrained distributed DWSC problem and user preferences are considered. For finding trade-off solutions for those problems, new Multi-objective Evolutionary Algorithms (MOEAs) are proposed based on the current Non-dominated Sorting Genetic Algorithm-II (NSGA-II). Furthermore, a new problem formulation for the dynamic distributed DWSC (D2−DWSC) problem with bandwidth fluctuations is proposed. An EC-based approach is developed to solve the D2-DWSC. Finally, extensive empirical evaluations are conducted that demonstrate the high performance of our proposed methods in finding composite services with good QoS.</p>

Download Full-text

The Impact of Distributed Data in Big Data Platforms on Organizations

Proceedings of the Future Technologies Conference (FTC) 2018 - Advances in Intelligent Systems and Computing ◽

10.1007/978-3-030-02683-7_76 ◽

2018 ◽

pp. 1024-1036 ◽

Cited By ~ 2

Author(s):

Oded Koren ◽

Matan Binyaminov ◽

Nir Perel

Keyword(s):

Big Data ◽

Distributed Data ◽

The Impact

Download Full-text

Pengaruh Gender, Kompleksitas Tugas, Tekanan Ketaatan, Pengalaman Auditor Terhadap Audit Judgement

Jurnal ULTIMA Accounting ◽

10.31937/akuntansi.v7i1.79 ◽

2015 ◽

Vol 7 (1) ◽

pp. 1-20

Author(s):

Stacia Pektra ◽

Ratnawati Kurnia

Keyword(s):

Task Complexity ◽

Hypothesis Test ◽

Distributed Data ◽

Test Results ◽

The Public ◽

Normality Test ◽

Audit Judgment ◽

Convenience Sampling ◽

The Impact ◽

Public Accountant

The purpose of this research was to examine the impact of gender, task complexity, obedience pressure, experience of auditors towards Audit Judgement. The object the auditors who works in the Public Accountant firms in Jakarta and Tangerang which have at least 3 years experience or have position as senior auditor. Data that had been analyzed were 110 questionnaires and the data were primer data. The type in this research is causal study and the sampling techniques that used is convenience sampling. The method that used is multiple regression analysis. The data that had been used from the entire questionnaires were valid. Reliability test results show questionnaires in this study is reliable. Normality test results showed all variables were normally distributed. Data used indicate the absence of the classical assumption of heteroscedasticity and non-occurrence of symptoms multicoloniarity between variables. Hypothesis test results indicate a strong correlation between variables and adjusted R-square value of 34.6%. In partial test only complexity of tasks that affect theentire audit judgment and all variables influential simultaneously. Keywords: Audit Experience, Audit Judgement, Gender, Obedience Pressure, Task Complexity

Download Full-text

Improving multi-class Boosting-based object detection

Integrated Computer-Aided Engineering ◽

10.3233/ica-200636 ◽

2020 ◽

Vol 28 (1) ◽

pp. 81-96

Author(s):

José Miguel Buenaposada ◽

Luis Baumela

Keyword(s):

Deep Learning ◽

Object Detection ◽

Data Augmentation ◽

Detection Performance ◽

Significant Progress ◽

Training Techniques ◽

Multi Scale ◽

Bounding Box ◽

Open Issue ◽

The Impact

In recent years we have witnessed significant progress in the performance of object detection in images. This advance stems from the use of rich discriminative features produced by deep models and the adoption of new training techniques. Although these techniques have been extensively used in the mainstream deep learning-based models, it is still an open issue to analyze their impact in alternative, and computationally more efficient, ensemble-based approaches. In this paper we evaluate the impact of the adoption of data augmentation, bounding box refinement and multi-scale processing in the context of multi-class Boosting-based object detection. In our experiments we show that use of these training advancements significantly improves the object detection performance.

Download Full-text

One-Bit Normalized Scatter Matrix Estimation For Complex Elliptically Symmetric Distributions

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) ◽

10.1109/icassp40776.2020.9053956 ◽

2020 ◽

Author(s):

Chun-Lin Liu ◽

P. P. Vaidyanathan

Keyword(s):

Symmetric Distributions ◽

Scatter Matrix ◽

Matrix Estimation ◽

Elliptically Symmetric Distributions

Download Full-text

Ruptured Status Discrimination Performance of Aspect Ratio, Height/Width, and Bottleneck Factor Is Highly Dependent on Aneurysm Sizing Methodology

Neurosurgery ◽

10.1227/neu.0b013e3182503bf9 ◽

2012 ◽

Vol 71 (1) ◽

pp. 38-46 ◽

Cited By ~ 37

Author(s):

Alexandra Lauric ◽

Merih I. Baharoglu ◽

Adel M. Malek

Keyword(s):

Aspect Ratio ◽

Statistical Significance ◽

Area Under The Curve ◽

Cerebral Aneurysms ◽

Discrimination Performance ◽

Average Diameter ◽

Detection Performance ◽

Width Ratio ◽

Neck Size ◽

The Impact

Abstract BACKGROUND: Numerous size and shape parameters have historically been used to describe cerebral aneurysms and to correlate rupture status. These parameters are often inconsistently defined. OBJECTIVE: To evaluate the impact of definition variation on rupture status detection performance. METHODS: Catheter rotational angiographic data sets of 134 consecutive aneurysms (60 ruptured) were automatically measured in 3 dimensions with a validated algorithm. According to the literature, aneurysm height was assessed as both maximal and orthogonal distances from dome to neck. Maximal and orthogonal widths were defined perpendicular to height definitions. Neck size was evaluated as minimum, maximum, and average diameter of the neck plane. Aspect ratio (AR; height/neck), height/width ratio (HW), and bottleneck factor (BNF; width/neck) were evaluated for alternative definitions of each size variable. Univariate statistics were used to identify significant features and to compute the area under the curve (AUC) of the receiver-operating characteristic. RESULTS: The AR, HW, and BNF showed significant dependence on parameter definition. Statistical significance and performance varied widely, depending on alternative definitions: AR, AUC range of 0.59 to 0.75; HW, AUC range of 0.48 to 0.72; and BNF, AUC range of 0.57 to 0.72. Using maximal height, orthogonal width, and minimum neck resulted in the best AR, HW, and BNF performances. Compared with HW, AR and BNF were less sensitive to alternative definitions. CONCLUSION: Alternative aneurysm size definitions have a significant impact on prediction performance and optimal threshold values. Adoption of standard methodology and sizing nomenclature appears critical to ensure rupture detection performance and reproducibility across studies.

Download Full-text

An efficient ensemble Kalman Filter implementation via shrinkage covariance matrix estimation: exploiting prior knowledge

Computational Geosciences ◽

10.1007/s10596-021-10035-4 ◽

2021 ◽

Vol 25 (3) ◽

pp. 985-1003

Author(s):

Santiago Lopez-Restrepo ◽

Elias D. Nino-Ruiz ◽

Luis G. Guzman-Reyes ◽

Andres Yarce ◽

O. L. Quintero ◽

...

Keyword(s):

Kalman Filter ◽

Prior Knowledge ◽

Covariance Matrix ◽

Ensemble Kalman Filter ◽

General Circulation ◽

Circulation Model ◽

Covariance Matrix Estimation ◽

Practical Implementation ◽

Matrix Estimation ◽

The Impact

AbstractIn this paper, we propose an efficient and practical implementation of the ensemble Kalman filter via shrinkage covariance matrix estimation. Our filter implementation combines information brought by an ensemble of model realizations, and that based on our prior knowledge about the dynamical system of interest. We perform the combination of both sources of information via optimal shrinkage factors. The method exploits the rank-deficiency of ensemble covariance matrices to provide an efficient and practical implementation of the analysis step in EnKF based formulations. Localization and inflation aspects are discussed, as well. Experimental tests are performed to assess the accuracy of our proposed filter implementation by employing an Advection Diffusion Model and an Atmospheric General Circulation Model. The experimental results reveal that the use of our proposed filter implementation can mitigate the impact of sampling noise, and even more, it can avoid the impact of spurious correlations during assimilation steps.

Download Full-text