Search

In this paper, a robust extreme learning machine is proposed. In comparison with the original extreme learning machine and the regularized extreme learning machine, this robust algorithm minimizes both the mean and variance of modeling errors in the objective function to overcome the bias-variance dilemma. As a result, its generalization performance and robustness are enhanced, and these merits are further proved theoretically. In addition, this proposed algorithm can keep the same computational efficiency as the original extreme learning machine and the regularized extreme learning machine. Then, several benchmark data sets are used to test the effectiveness and soundness of the proposed algorithm. Finally, it is employed to model a real small-scale turbojet engine. This engine is fit well. Especially, on the idle phase, where the signal-to-noise ratio is low and it is very hard to model, the proposed algorithm performs well and its robustness is sufficiently showcased. All in all, the proposed algorithm provides a candidate technique for modeling real systems.

Quorum sensing (QS) is a bacterial communication system that involves production and sensing of extracellular signals. In laboratory models, QS allows bacteria to monitor and respond to their own cell density and is critical for fitness. However, how QS proceeds in natural, spatially structured bacterial communities is not well understood, which significantly hampers our understanding of the emergent properties of natural communities. To address this gap, we assessed QS signaling in the opportunistic pathogen Pseudomonas aeruginosa in a cystic fibrosis (CF) lung infection model that recapitulates the biogeographical aspects of the natural human infection. In this model, P. aeruginosa grows as spatially organized, highly dense aggregates similar to those observed in the human CF lung. By combining this natural aggregate system with a micro-3D-printing platform that allows for confinement and precise spatial positioning of P. aeruginosa aggregates, we assessed the impact of aggregate size and spatial positioning on both intra- and interaggregate signaling. We discovered that aggregates containing ∼2,000 signal-producing P. aeruginosa were unable to signal neighboring aggregates, while those containing ≥5,000 cells signaled aggregates as far away as 176 µm. Not all aggregates within this "calling distance" responded, indicating that aggregates have differential sensitivities to signal. Overexpression of the signal receptor increased aggregate sensitivity to signal, suggesting that the ability of aggregates to respond is defined in part by receptor levels. These studies provide quantitative benchmark data for the impact of spatial arrangement and phenotypic heterogeneity on P. aeruginosa signaling in vivo.

This article outlines the design and delivery of sessions used and the results obtained from a 2016 pilot study investigating whether the use of mindfulness-based techniques could enhance the wellbeing and academic performance of university students. The study, a collaboration between counselling and chaplaincy at Charles Sturt University (CSU) had three aims: (1) to determine benchmark data of wellbeing, focused attention and academic motivation across the CSU student population; (2) to determine whether mindfulness training affected the emotional wellbeing, focused attention and academic performance of higher education students; and (3) because around 60% of CSU students study online, to examine the relative contribution of two different online modes for delivering mindfulness training, one in real time and one self-guided. Based on research that supports the assumption that ‘mindful learning’ benefits students by enhancing their cognitive and socio-emotional capabilities and improving their general wellbeing and academic performance, this study used three validated psychometric scales to measure the attention and awareness, wellbeing and academic motivation of the participants before and after a four week mindfulness training course. The results demonstrated three main findings: mindfulness training may contribute in the long-term to creating a mindset conducive to learning; mindfulness training can be successfully administered online, both in real time and self-guided; and mindfulness training may contribute to shifting patterns of motivation for learning from extrinsic to intrinsic motivation.

The concept of `diversity' has been one of the main open issues in the field of multiple classifier systems. In this paper we address a facet of diversity related to its effectiveness for ensemble construction, namely, explicitly using diversity measures for ensemble construction techniques based on the kind of overproduce and choose strategy known as ensemble pruning. Such a strategy consists of selecting the (hopefully) more accurate subset of classifiers out of an original, larger ensemble. Whereas several existing pruning methods use some combination of individual classifiers' accuracy and diversity, it is still unclear whether such an evaluation function is better than the bare estimate of ensemble accuracy. We empirically investigate this issue by comparing two evaluation functions in the context of ensemble pruning: the estimate of ensemble accuracy, and its linear combination with several well-known diversity measures. This can also be viewed as using diversity as a regularizer, as suggested by some authors. To this aim we use a pruning method based on forward selection, since it allows a direct comparison between different evaluation functions. Experiments on thirty-seven benchmark data sets, four diversity measures and three base classifiers provide evidence that using diversity measures for ensemble pruning can be advantageous over using only ensemble accuracy, and that diversity measures can act as regularizers in this context.

This paper concerns one of intelligent computational techniques which is case based reasoning (CBR) more particularly the case base maintenance (CBM). It aims to ensure the CBR systems quality. Throughout this paper, we were faced to a problematic question: how to shrink the size of the case base while preserving as much as possible the performance and the competence of the CBR system in soft context. To answer this question, we have first analyzed and revised the theoretical foundations of the existing CBM methods. Then, we have proposed a novel soft case base maintenance (SCBM) method based on a soft competence model (SCM) and a fuzzy clustering technique. Our method has the objective to guarantee the CBR systems efficiency in terms of improving the competence, and reducing both the storage requirements and search time. We support our approach with empirical evaluation using different benchmark data sets to show the effectiveness of our method in terms of improving the competence and the performance of the system.

Purpose Publication metrics are useful in evaluating academic faculty for awarding grants, recruitment, and promotion. A new metric, the relative citation ratio (RCR), was recently released by the National Institutes of Health (NIH); however, no benchmark data yet exist. We sought to create benchmark data for physician faculty in academic radiation oncology (RO) and analyze correlations associated with increased academic productivity. Methods Citation database searches were performed for all US radiation oncologists affiliated with academic RO programs. Gender, NIH funding, career duration, academic rank, RCR, and weighted RCR were collected for each faculty. RCR and weighted RCR were calculated and compared between each subgroup of interest. RCR percentiles were also created for reference. Results A total of 1,299 RO physician faculty members from 75 institutions were included in the analysis. Overall, RO physician were very productive and influential with a mean RCR of 1.57 ± 1.53 SD and median RCR (interquartile range) of 1.32 (0.87-1.94). Academic rank, career duration, and NIH funding were associated with increased mean RCR and weighted RCR. Male gender and having a PhD were associated with an increased weighted RCR but not an increased mean RCR. Conclusions Current academic radiation oncologists have a high mean RCR value relative to the benchmark NIH RCR value of 1. All subgroups analyzed had an RCR value above 1 with professor or chair and previous NIH funding having the highest RCR and weighted RCR values overall. These data may be useful for self-evaluation of ROs as well as evaluation of faculty by institutional and departmental leaders.

In this paper, we propose a method for solving the twin bounded support vector machine (TBSVM) for the binary classification. To do so, we use the augmented Lagrangian (AL) optimization method and smoothing technique, to obtain new unconstrained smooth minimization problems for TBSVM classifiers. At first, the augmented Lagrangian method is recruited to convert TBSVM into unconstrained minimization programming problems called as AL-TBSVM. We attempt to solve the primal programming problems of AL-TBSVM by converting them into smooth unconstrained minimization problems. Then, the smooth reformulations of AL-TBSVM, which we called AL-STBSVM, are solved by the well-known Newton's algorithm. Finally, experimental results on artificial and several University of California Irvine (UCI) benchmark data sets are provided along with the statistical analysis to show the superior performance of our method in terms of classification accuracy and learning speed.

Satellite-based precipitation products (SPP) have been the focus of much attention for their potential utility of drought monitoring and analysis. Assessing the accuracy and drought performance of the SPPs for ungauged areas is urgent and also challenging. This study comprehensively investigated the performance of the triple collocation (TC) approach, an assessment approach requiring no benchmark data, the accuracy and drought utility assessment of the SPPs. Furthermore, we applied the TC approach experimentally to the SPPs throughout mainland China. The China monthly Precipitation Analysis Product (CPAP), a gridded gauge-based precipitation dataset, was employed to validate the TC approach. Results demonstrate the effectiveness of the TC approach in retrieving spatial patterns in the accuracy and drought performance (via short-timescale drought indices) of the SPPs, but systematic bias was observed. Moreover, the TC approach exhibits a weak performance in assessing the long-timescale drought indices. By comparing the assessment results, the TC approach reveals the potential underestimation of the SPP accuracy by the traditional assessment approach for some ungauged regions. In general, the proposed TC outperforms the traditional approach in qualitatively assessing the performance pattern of SPPs over ungauged and poorly gauged areas, but the systematic bias of TC needs attention when quantitatively determining SPP performance.

Next generation sequencing is a standard tool used in clinical diagnostics. In Mendelian diseases the challenge is to discover the single etiological variant among thousands of benign or functionally unrelated variants. After calling variants from aligned sequencing reads, variant prioritisation tools are used to examine the conservation or potential functional consequences of variants. We hypothesised that the performance of variant prioritisation tools may vary by disease phenotype. To test this we created benchmark data sets for variants associated with different disease phenotypes. We found that performance of 24 tested tools is highly variable and differs by disease phenotype. The task of identifying a causative variant amongst a large number of benign variants is challenging for all tools, highlighting the need for further development in the field. Based on our observations, we recommend use of five top performers found in this study (FATHMM, M-CAP, MetaLR, MetaSVM and VEST3). In addition we provide tables indicating which analytical approach works best in which disease context. Variant prioritisation tools are best suited to investigate variants associated with well-studied genetic diseases, as these variants are more readily available during algorithm development than variants associated with rare diseases. We anticipate that further development into disease focussed tools will lead to significant improvements., Genomic analysis: Tools for prioritizing gene variants impacted by disease context The performance of software tools used to distinguish disease-causing genetic variants depends on the type of disease under investigation. Denise Anderson and Timo Lassmann from the Telethon Kids Institute in Subiaco, Australia, compared 24 software tools commonly used to narrow down DNA variants found in next-generation sequencing data to those likely to cause a particular disease. By looking at more than 4000 disease phenotypes, the researchers found that the different prioritisation tools, owing to their different methodologies and algorithms, varied in their ability to discriminate between pathogenic and benign gene variants. The top-performing tools all used machine-learning techniques and worked best in cases of well-studied genetic diseases. The findings highlight the need for additional disease-focused tool development, and offer a resource to help researchers decide which approach to use in different disease contexts.

Colocalization is a powerful tool to study the interactions between fluorescently labeled molecules in biological fluorescence microscopy. However, existing techniques for colocalization analysis have not undergone continued development especially in regards to robust statistical support. In this paper, we examine two of the most popular quantification techniques for colocalization and argue that they could be improved upon using ideas from nonparametric statistics and scan statistics. In particular, we propose a new colocalization metric that is robust, easily implementable, and optimal in a rigorous statistical testing framework. Application to several benchmark data sets, as well as biological examples, further demonstrates the usefulness of the proposed technique.

Natural image quality assessment (NIQA) wins increasing attention, while NIQA models are rarely used in the medical community. A couple of studies employ the NIQA methodologies for medical image quality assessment (MIQA), but building the benchmark data sets necessitates considerable time and professional skills. In particular, the characteristics of synthesized distortions are different from those of clinical distortions, which make the results not so convincing. In clinic, signal-to-noise ratio (SNR) is widely used, which is defined as the quotient of the mean signal intensity measured in a tissue region of interest (ROI) and the standard deviation of the signal intensity in an air region outside the imaged object, and both regions are outlined by specialists. We take advantage of the knowledge that SNR is routinely used and concern whether SNR measure can perform as a baseline metric for the development of MIQA algorithms. To address the issue, the inter-observer reliability of SNR measure is investigated regarding to different tissue ROIs [white matter (WM); cerebral spinal fluid (CSF)] in magnetic resonance (MR) images. A total of 192 ${\text{T}}_{2}^{*}$ , 88 ${\text{T}}_{1}$ , 76 ${\text{T}}_{2}$ and 55 contrast-enhanced ${\text{T}}_{1}$ ( ${\text{T}}_{1}\text{C}$ ) weighted images are analyzed. Statistical analysis indicates that SNR values show consistency between different observers to the same ROI in each modality (Wilcoxon rank sum test, $p_{w}\ge 0.11$ ; and paired sample $t$ -test, $p_{p}\ge 0.28$ ). Moreover, whether off-the-shelf NIQA models can predict MR image quality is considered by using SNR values as the reference scores. Four NIQA models (BIQI, BLIINDS-II, BRISQUE, and NIQE) are evaluated, and the correlation between SNR values and NIQA results is evaluated. Pearson correlation coefficient ( $r_{p}$ ) shows that WM-based SNR values correlates well with BIQI, BLIINDS-II and BRISQUE in ${\text{T}}_{2}^{*}$ images ( $r_{p}\geq 0.77$ ), BRISQUE and NIQE in ${\text{T}}_{1}$ images ( $r_{p}\geq 0.75$ ), BLIINDS-II in ${\text{T}}_{2}$ images ( $r_{p}\geq 0.67$ ), and BRISQUE and NIQE in ${\text{T}}_{1}\text{C}$ images ( $r_{p}\geq 0.58$ ), while CSF-based SNR values correlates well with BLIINDS-II in ${\text{T}}_{2}^{*}$ images ( $r_{p}\geq 0.64$ ) and ${\text{T}}_{2}$ images ( $r_{p}\geq 0.60$ ), and all $p_{p} . The prediction performance analysis further proves the result from the correlation analysis. Conclusively, SNR measure is reliable to different observations and can perform as a baseline indicator for the development of MIQA algorithms. In general, BRISQUE and BLIINDS-II are full of potential to be conditionally used as objective MIQA models toward human brain MR images. This paper presents the first attempt of using SNR measure to bridge the gap between NIQA and MIQA, and large-scale experiments should be further conducted to confirm the conclusion in this paper.

This paper introduces the EREL algorithm that integrates Entity Recognition, Co-reference Resolution (CR) and Disambiguation. The algorithm recognizes entity mentions as the longest name based on the name dictionary constructed from the Wikipedia data. The CR is integrated into the algorithm to improve the performance in processing short-form or abbreviated names. The algorithm employs a new approach in disambiguation entities using new features as entity-level context information and case-sensitive data about the mention in disambiguation. Tested on four benchmark data sets in the GERBIL framework, EREL outperforms the current Entity Linking methods. EREL achieves the micro f-score as 0.83 in both tasks Disambiguate to Wikipedia and Annotate to Wikipedia.

To characterize and compare the experiences of matched applicants and program directors (PDs) participating in the first large-scale integrated interventional radiology (IR) residency match.Survey questionnaires were distributed nationally to integrated IR applicants who matched in the 2017 Match cycle and PDs. Both groups were questioned regarding their experiences with the application, interview, rank, and match processes as well as applicant-specific and PD-specific information. Summary and descriptive statistics were applied to responses, and comparison of Likert scale responses was performed by two-sample t test.Sixty-one matched applicants (51.3%) and 34 PDs (55.7%) responded to the survey. Regarding the match process, applicants believed United States Medical Licensing Examination (USMLE) Step 1 score (P = .002) and connection to a program's geographic location (P = .006) were significantly more important than PDs did, whereas PDs ranked grades (P = .049), class rank (P = .011), academic awards (P = .003), additional degrees (P.001), and USMLE Step 2 Clinical Skills score (P.001) as significantly more important factors than applicants did. Additional information regarding demographic data, medical school experiences in IR, application strategies, interview experiences, rank lists, the intern year, and match results are reported.The completion of the first large-scale integrated IR match represents a paradigm shift in the way in which IR practitioners are recruited and trained. This study provides valuable benchmark data and analysis that can be used to improve efforts to match the best-fitting applicants into the integrated IR residency and improve future match cycles for applicants and PDs alike.

While some organizations swear by the benefits of transparency and are eager to know how to implement it, many managers are still reluctant or even afraid to use it. Our research reveals that only a few innovative companies have taken steps to leverage a potentially useful form of transparency: the provision of accessible and objective information to customers (e.g., by sharing unbiased benchmark data, publishing unfiltered customer comments, or providing candid product reviews that may praise but also criticize the company’s products). Our study also shows that that many companies remain wary and view greater calls for transparency as a challenge to be managed rather than an opportunity to be traded upon. This is partly due to limited research into the performance benefits of giving customers access to objective information, and lack of practical guidelines on how to actually implement it. The current study addresses these shortcomings: first, by investigating whether performance transparency leads to customer outcomes that can be profitable for an organization; and second, by analyzing the characteristics of successful transparency initiatives in a wide range of industries. The research shows that customers exhibit more trust and are willing to pay a premium to deal with transparent businesses. Also, it uncovers seven effective approaches to implementing transparency. The key contributions of the study are therefore the provision of convincing empirical evidence about the benefits of performance transparency and ways in which management may implement it successfully.

Methods for improving learning accuracy by utilizing a plurality of data sets having different reliabilities have been extensively studied. Unreliable data sets often include data with incorrect labels, and it is known that learning from such datasets adversely affects learning accuracy. In our study we focused on the learning problem, Taming which deals with two kinds of data sets with different reliabilities. We propose a label estimation method for use in data sets that include the data with incorrect labels. The proposed method is an extension of BaggTaming, which is proposed as a solution to Taming. We conducted experiments to verify the effectiveness of the proposed method using a benchmark data set which included the data with intentionally changed labels to incorrect. We confirmed that the learning accuracy can be improved by the proposed method when learning by using the data sets with modified labels.

The kinetics of many reactions are critically dependent upon the barrier heights for which accurate determination can be difficult. From the perspective of attaining such quantities using computational quantum chemistry, it is important to appropriately validate routine and efficient methodologies such as density functional theory (DFT) procedures. In the present study, we embark on the journey of establishing diverse databases using a consistent high-level quantum chemistry procedure, against which new and existing methodologies can be assessed. Thus, we have used the composite protocol W3X-L to provide more than 100 refined reference values for existing databases [e.g., Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2005, 109, 5656] and additionally establish benchmark data that are of interest to atmospheric and combustion chemists. While our endeavor has just begun, assessment of various DFT methods with our existing results lends support to the use of MN15 as an adequate method for general kinetics applications. We also recommend the use of less-costly W2X and WG composite protocols for obtaining adequately accurate reference thermochemical values for larger molecular systems., Physical Chemistry Chemical Physics, 20(16), pp.10732-10740; 2018

Multidimensional Scaling (MDS) has been exploited to visualise the hidden structures among a set of entities in a reduced dimensional metric space. Here, we are interested in cases whenever the initial dissimilarity matrix is contaminated by outliers. It is well-known that the state-of-the-art algorithms for solving the MDS problem generate erroneous embeddings due to the distortion introduced by such outliers. To remedy this vulnerability, a unified framework for the solution of MDS problem is proposed, which resorts to half-quadratic optimization and employs potential functions of M-estimators in combination with l2,1 norm regularization. Two novel algorithms are derived. Their performance is assessed for various M-estimators against state-of-the-art MDS algorithms on four benchmark data sets. The numerical tests demonstrate that the proposed algorithms perform better than the competing alternatives.

The two-sided assembly line becomes very popular in recent years. In this paper, a priority rules-based algorithmic design is developed for optimizing two-sided assembly line. Five elementary rules and 90 composite rules are tested on the benchmark data sets and their performance are provided. Two enumerative principles, which are specific to two-sided assembly lines are proposed to enhance the performance of the rules. Further, priority rules are embedded into a bounded dynamic programming framework to form a deterministic algorithm where the use of a bound can reduce the solution space as the algorithm is advanced stage-by-stage. These approaches offer distinct advantages over the methods proposed in the literature, such as less fine-tuning effort and more stable results. Computational results show that the novel algorithm can generate good solutions efficiently, especially in large sized problems.

Integrative modeling approaches attempt to combine experiments and computation to derive structure-function relationships in complex molecular assemblies. Despite their importance for the advancement of life sciences, benchmarking of existing methodologies is rather poor. The 12th round of the Critical Assessment of protein Structure Prediction (CASP) offered a unique niche to benchmark data and methods from two kinds of experiments often used in integrative modeling, namely residue-residue contacts obtained through crosslinking/mass-spectrometry (CLMS), and small-angle X-ray scattering (SAXS) experiments. Upon assessment of the models submitted by predictors for 3 targets assisted by CLMS data and 11 targets by SAXS data, we observed no significant improvement when compared to the best data-blind models, although most predictors did improve relative to their own data-blind predictions. Only for target Tx892 of the CLMS-assisted category and for target Ts947 of the SAXS-assisted category, there was a net, albeit mild, improvement relative to the best data-blind predictions. We discuss here possible reasons for the relatively poor success, which point rather to inconsistencies in the data sources rather than in the methods, to which a few groups were less sensitive. We conclude with suggestions that could improve the potential of data integration in future CASP rounds in terms of experimental data production, methods development, data management and prediction assessment.

A knowledge graph is a graph with entities of different types as nodes and various relations among them as edges. The construction of knowledge graphs in the past decades facilitates many applications, such as link prediction, web search analysis, question answering, and so on. Knowledge graph embedding aims to represent entities and relations in a large-scale knowledge graph as elements in a continuous vector space. Existing methods, for example, TransE, TransH, and TransR, learn the embedding representation by defining a global margin-based loss function over the data. However, the loss function is determined during experiments whose parameters are examined among a closed set of candidates. Moreover, embeddings over two knowledge graphs with different entities and relations share the same set of candidates, ignoring the locality of both graphs. This leads to the limited performance of embedding related applications. In this article, a locally adaptive translation method for knowledge graph embedding, called TransA, is proposed to find the loss function by adaptively determining its margin over different knowledge graphs. Then the convergence of TransA is verified from the aspect of its uniform stability. To make the embedding methods up-to-date when new vertices and edges are added into the knowledge graph, the incremental algorithm for TransA, called iTransA, is proposed by adaptively adjusting the optimal margin over time. Experiments on four benchmark data sets demonstrate the superiority of the proposed method, as compared to the state-of-the-art ones.

Since the original extreme learning machine (ELM) generates the hidden nodes randomly, it usually needs more hidden nodes to reach the good classification performance. However, more hidden nodes will jeopardize the real time, which limits its applications to the testing time sensitive scenarios. To this end, the commonly-used methods tend to compact its structure via optimizing the number of hidden nodes. Different from this viewpoint of network structure, in this paper two algorithms are proposed to improve the real time performance of ELM from a viewpoint of data structure. Specially, they improve the ELM classification performance by retargeting its label vectors. As thus, they need fewer hidden nodes to reach the same classification performance, which means the better real time. Finally, experimental results on the benchmark data sets validate the effectiveness and feasibility of the presented two algorithms. To be more important, they are applied to the fault diagnosis of aircraft engine and can be developed as its candidate techniques.

How does a quasicrystal grow? Despite the decades of research that have been dedicated to this area of study, it remains one of the fundamental puzzles in the field of crystal growth. Although there has been no lack of theoretical studies on quasicrystal growth, there have been very few experimental investigations with which to test their various hypotheses. In particular, evidence of the in situ and three-dimensional (3D) growth of a quasicrystal from a parent liquid phase is lacking. To fill-in-the-gaps in our understanding of the solidification and melting pathways of quasicrystals, we performed synchrotron-based X-ray imaging experiments on a decagonal phase with composition of Al-15at%Ni-15at%Co. High-flux X-ray tomography enabled us to observe both growth and melting morphologies of the 3D quasicrystal at temperature. We determined that there is no time-reversal symmetry upon growth and melting of the decagonal quasicrystal. While quasicrystal growth is predominantly dominated by the attachment kinetics of atomic clusters in the liquid phase, melting is instead barrier-less and limited by buoyancy-driven convection. These experimental results provide the much-needed benchmark data that can be used to validate simulations of phase transformations involving this unique phase of matter.

The identically and independently distributed (i.i.d.) condition required by conventional machine learning approaches may sometimes be violated when patterns occur as groups (where each group shares a homogeneous style, called a field). By breaking it, we extend in this paper the famous Support Vector Machine (SVM) to a novel framework named Field Support Vector Machine (F-SVM), in which the training and predicting a group of patterns (i.e., a field pattern) are performed simultaneously. Specifically, the proposed F-SVM is learned by optimizing simultaneously both the classifier and the Style Normalization Transformation (SNT) for each group of data, even feasible in the high-dimensional kernel space. The SNT transform the original style-discriminative patterns to style-free ones, satisfying the i.i.d. assumption required by the conventional SVM learning and implementation. An efficient optimization algorithm is further developed with the convergence guaranteed theoretically. More importantly, by appropriately exploring the style consistency in each field, the proposed F-SVM model is able to significantly improve the classification accuracy. A series of experiments are conducted to verify the effectiveness and confirmed improvement on the performance of the F-SVM model. Empirical results show that the proposed F-SVM outperforms other relevant baselines in two different benchmark data sets.

This paper presents results obtained by reproducing a VVER-1000 lattice physics benchmark using the Serpent2 code. The benchmark data are related to the initial cycles of the Khmelnitskiy nuclear power plant. This study compares results obtained with the Serpent2 code against those calculations published in the benchmark document. Detailed descriptions are provided for two fuel assemblies; one without gadolinium in the fuel pins, and the other with gadolinium. A sensitivity study is performed utilizing different nuclear data libraries, different depletion sequences (including pin by pin burnup) and different number of radial regions within the fuel pins containing gadolinium. The infinite multiplication factor behavior and isotopic concentrations of major isotopes are compared and discussed. It is concluded that Serpent2 provided reasonable results in comparison with the other codes used in the benchmark.

In this paper, we propose a novel matching method to establish dense correspondences automatically between two images in a hierarchical superpixel-to-pixel manner. Our method first estimates dense superpixel pairings between the two images in the coarse-grained level to overcome large patch displacements and then utilizes superpixel level pairings to drive the matchings in the pixel level to obtain fine texture details. In order to compensate for the influence of color and illumination variations, we apply a regularization technique to rectify images by a color transfer function. Experimental validation on benchmark data sets demonstrates that our approach achieves better visual quality outperforming the state-of-the-art dense matching algorithms.

A practical mathematical programming based approach is introduced for solving the examination timetabling problem at the German Jordanian University (GJU), whereby the complex process of acquiring a feasible examination timetable is simplified by subdividing it into three smaller sub-problems (phases). Accordingly, the exams are initially allocated to time slots in phase one, the time slots are then allotted to days in phase two, and finally in phase three the exams are assigned to rooms based on the number of students taking each exam and capacities of the rooms. The solution for each phase is acquired based on an integer linear programming (ILP) formulation, while satisfying a set of hard constraints that ensure comfortable exam timetables for all students and meet the desired requirements set by GJU administrative staff. Furthermore, the solver can be controlled and launched from a student information system named MyGJU Admin, which enabled registrars at the university to easily, quickly, and accurately generate final exam timetables in several standard formats. Moreover, the approach was validated based on recent GJU registration information as well as real-world benchmark data.

Discontinuity enhancement attributes are commonly used to facilitate the interpretation process by enhancing edges in seismic images and providing a quantitative measure of the significance of discontinuous features. These attributes require careful preprocessing to maintain geologic features and suppress acquisition and processing artifacts, which may be artificially detected as a geologic edge. We have developed the plane-wave Sobel attribute, a modification of the classic Sobel filter, by orienting the filter along seismic structures using plane-wave destruction and plane-wave shaping. The plane-wave Sobel attribute can be applied directly to a seismic image to efficiently and effectively enhance discontinuous features or to a coherence image to create a sharper and more detailed image. Two field benchmark data examples with many faults and channel features from offshore New Zealand and offshore Nova Scotia demonstrate the effectiveness of this method compared with conventional coherence attributes. The results are reproducible using the Madagascar software package.

Dial-a-ride services provide collective on-demand transportation, usually tailored to the needs of people with reduced mobility. This paper investigates the operational effects of horizontal cooperation among dial-a-ride providers. The current practice is that users choose a particular provider to submit their requests. Providers operating in the same area create their routing solutions independently of each other, given their own set of customers. In contrast, horizontal cooperation through joint route planning implies that customer requests can be exchanged among providers in order to minimize the overall routing cost. In addition to quantifying the operational benefits generated by such a horizontal cooperation, this paper identifies operational characteristics that influence the magnitude of the savings. A real-life case study reveals the reasons why providers benefit from certain request exchanges, as well as the extent to which these exchanges are predictable in advance. The solutions are obtained using a large neighborhood search algorithm that performs well on benchmark data.

In many real life applications, a complete labeled data set is not always available. Therefore, an ideal learning algorithm should be able to learn from both labeled and unlabeled data. In this work a two stage local boosting algorithm for handling semi-supervised classification tasks is proposed. The proposed method can be simply described as: (a) a two stage local boosting method, (b) which adds self-labeled examples of unlabeled data and (c) employ them on semi-supervised classification tasks. Grounded on the local application of the boosting-by-reweighting version of AdaBoost, the proposed method utilizes unlabeled data to enhance it’s classification performance. Simulations on thirty synthetic and real-world benchmark data sets show that the proposed method significantly outperforms nine other well-known semi-supervised classification methods in terms of classification accuracy.

We introduce and analyze various approaches for computing excited electronic states using our recently developed adaptive configuration interaction (ACI) method [ Schriber , J. B. and Evangelista , F. A. J. Chem. Phys. 2016 , 144 , 161106 ]. These ACI methods aim to describe multiple electronic states with equal accuracy, including challenging cases like multielectron, charge-transfer, and near-degenerate states. We develop both state-averaged and state-specific approaches to compute excited states whose absolute energy error can be tuned by a user-specified energy error threshold, σ. State-averaged schemes are found to be more efficient in that they obtain all of the states simultaneously in one computation, but they lose some degree of statewise tunability. State-specific algorithms allow for direct control of the error of each state, though the states must be computed sequentially. We compare each method using methylene, LiF, and all-trans polyene benchmark data.

This landmark study of doctoral graduates in mathematics education provides benchmark data that we hope will be useful to the mathematics education community. Five hundred three doctoral gr...

The ability to interpret the predictions made by quantitative structure-activity relationships (QSARs) offers a number of advantages. While QSARs built using nonlinear modeling approaches, such as the popular Random Forest algorithm, might sometimes be more predictive than those built using linear modeling approaches, their predictions have been perceived as difficult to interpret. However, a growing number of approaches have been proposed for interpreting nonlinear QSAR models in general and Random Forest in particular. In the current work, we compare the performance of Random Forest to those of two widely used linear modeling approaches: linear Support Vector Machines (SVMs) (or Support Vector Regression (SVR)) and partial least-squares (PLS). We compare their performance in terms of their predictivity as well as the chemical interpretability of the predictions using novel scoring schemes for assessing heat map images of substructural contributions. We critically assess different approaches for interpreting Random Forest models as well as for obtaining predictions from the forest. We assess the models on a large number of widely employed public-domain benchmark data sets corresponding to regression and binary classification problems of relevance to hit identification and toxicology. We conclude that Random Forest typically yields comparable or possibly better predictive performance than the linear modeling approaches and that its predictions may also be interpreted in a chemically and biologically meaningful way. In contrast to earlier work looking at interpretation of nonlinear QSAR models, we directly compare two methodologically distinct approaches for interpreting Random Forest models. The approaches for interpreting Random Forest assessed in our article were implemented using open-source programs that we have made available to the community. These programs are the rfFC package ( https://r-forge.r-project.org/R/?group_id=1725 ) for the R statistical programming language and the Python program HeatMapWrapper [ https://doi.org/10.5281/zenodo.495163 ] for heat map generation.

In this study, a novel total spectrum-based non-gray model is proposed by modifying the narrow band-based WSGG regrouping (RG-NBW) method. Using the total weighting factor and the total absorption coefficient for each gray gas group, the required number to solve the RTE can be remarkably reduced. To verify the performance of the modified RG-NBW, three-dimensional radiative transfer problems using pure H 2 O gas with a uniform temperature, a non-homogeneous mixture of H 2 O and N 2 with a uniform temperature, and a mixture of CO 2 , H 2 O and N 2 with a non-uniform temperature were solved using the discrete ordinate method (DOM) with various non-gray models. The results are compared to benchmark data obtained using the ray-tracing method with the statistical narrow band (SNB) model. For all cases, the results with the modified RG-NBW model show good agreement with the benchmark results as well as results from other models. Furthermore, the modified RG-NBW model requires the shortest computational time after the WSGG model. Therefore, considering the accuracy and computational time, the modified RG-NBW model can be an efficient choice for non-gray gas radiation problems.

Shape characterisation is important in many fields dealing with spatial data. For this purpose, numerous shape analysis and recognition methods with different degrees of complexity have so far been developed. Among them, relatively simple indices are widely used in spatial applications, but their performance has not been investigated sufficiently, particularly for building footprints (BFs). Therefore, this article focuses on BF shape characterisation with shape indices and classification schemes in a GIS environment. This study consists of four phases. In the first phase, the criteria for BF shape complexity were identified, and accordingly, benchmark data was constructed by human experts in three shape complexity categories. In the second phase, 18 shape indices were selected from the literature and automatically computed in GIS. The performance of these indices was then statistically assessed with histograms, correlation matrix and boxplots, and consequently four indices were found to be appropr...

A recent technology which makes possible for us to interact with automated systems without using any body part is called Brain Computer Interface (BCI). In its concrete applications, electroencephalogram (EEG) is benefited by a BCI environment for being capable of obtaining brain waves. In our study, evaluation of success rates of the predictions made by C x k - Nearest Neighborhood (Cxk-NN) Algorithm for EEG Eye State Data whose states are called “Opened Eye“ and “Closed Eye“ is applied. This EEG Eye State dataset is obtained from UCI Machine Learning Repository on the web and it is a highly-used benchmark data on this field. As there are only two classes of the signals, we test binary classification performance of our classification algorithm (Cxk –NN). Comparison of those values with the ones obtained by the other successful classification algorithms in the literature applied on the same data set also take place in our study. Cxk-NN is an instance-based classification method advanced from simple k – Nearest Neighborhood Algorithm, and improved success results are observed when it is compared with k-NN. Keywords: brain computer interface (bci), classificaiton, eeg, c x k - nearest neighborhood algorihtm.

The IntaRNA algorithm enables fast and accurate prediction of RNA–RNA hybrids by incorporating seed constraints and interaction site accessibility. Here, we introduce IntaRNAv2, which enables enhanced parameterization as well as fully customizable control over the prediction modes and output formats. Based on up to date benchmark data, the enhanced predictive quality is shown and further improvements due to more restrictive seed constraints are highlighted. The extended web interface provides visualizations of the new minimal energy profiles for RNA–RNA interactions. These allow a detailed investigation of interaction alternatives and can reveal potential interaction site multiplicity. IntaRNAv2 is freely available (source and binary), and distributed via the conda package manager. Furthermore, it has been included into the Galaxy workflow framework and its already established web interface enables ad hoc usage.

This is the last paper in the Synthesis section of this special issue on `Same Data, Different Results'. We first provide a framework of how to describe and distinguish approaches to topic extraction from bibliographic data of scientific publications. We then compare solutions delivered by the different topic extraction approaches in this special issue, and explore where they agree and differ. This is achieved without reference to a ground truth, since we have to assume the existence of multiple, equally important, valid perspectives and want to avoid bias through the adoption of an ad-hoc yardstick. Instead, we apply different ways to quantitatively and visually compare solutions to explore their commonalities and differences and develop hypotheses about the origin of these differences. We conclude with a discussion of future work needed to develop methods for comparison and validation of topic extraction results, and express our concern about the lack of access to non-proprietary benchmark data sets to support method development in the field of scientometrics.

Theoreticians and experimentalists should work together more closely to establish reliable rankings and benchmarks for quantum chemical methods. Comparison to carefully designed experimental benchmark data should be a priority. Guidelines to improve the situation for experiments and calculations are proposed.

Robust principal component analysis (PCA) is one of the most important dimension-reduction techniques for handling high-dimensional data with outliers. However, most of the existing robust PCA presupposes that the mean of the data is zero and incorrectly utilizes the average of data as the optimal mean of robust PCA. In fact, this assumption holds only for the squared [Formula: see text]-norm-based traditional PCA. In this letter, we equivalently reformulate the objective of conventional PCA and learn the optimal projection directions by maximizing the sum of projected difference between each pair of instances based on [Formula: see text]-norm. The proposed method is robust to outliers and also invariant to rotation. More important, the reformulated objective not only automatically avoids the calculation of optimal mean and makes the assumption of centered data unnecessary, but also theoretically connects to the minimization of reconstruction error. To solve the proposed nonsmooth problem, we exploit an efficient optimization algorithm to soften the contributions from outliers by reweighting each data point iteratively. We theoretically analyze the convergence and computational complexity of the proposed algorithm. Extensive experimental results on several benchmark data sets illustrate the effectiveness and superiority of the proposed method.

BACKGROUND: Many functional analysis tools have been developed to extract functional and mechanistic insight from bulk transcriptome data. With the advent of single-cell RNA sequencing (scRNA-seq), it is in principle possible to do such an analysis for single cells. However, scRNA-seq data has characteristics such as drop-out events and low library sizes. It is thus not clear if functional TF and pathway analysis tools established for bulk sequencing can be applied to scRNA-seq in a meaningful way. RESULTS: To address this question, we perform benchmark studies on simulated and real scRNA-seq data. We include the bulk-RNA tools PROGENy, GO enrichment, and DoRothEA that estimate pathway and transcription factor (TF) activities, respectively, and compare them against the tools SCENIC/AUCell and metaVIPER, designed for scRNA-seq. For the in silico study, we simulate single cells from TF/pathway perturbation bulk RNA-seq experiments. We complement the simulated data with real scRNA-seq data upon CRISPR-mediated knock-out. Our benchmarks on simulated and real data reveal comparable performance to the original bulk data. Additionally, we show that the TF and pathway activities preserve cell type-specific variability by analyzing a mixture sample sequenced with 13 scRNA-seq protocols. We also provide the benchmark data for further use by the community. CONCLUSIONS: Our analyses suggest that bulk-based functional analysis tools that use manually curated footprint gene sets can be applied to scRNA-seq data, partially outperforming dedicated single-cell tools. Furthermore, we find that the performance of functional analysis tools is more sensitive to the gene sets than to the statistic used. CHH is supported by the German Federal Ministry of Education and Research (BMBF)-funded project Systems Medicine of the Liver (LiSyM, FKZ: 031 L0049). MPK, BAJ, and DAL are supported by NIH Grant U54-CA217377. BS is supported by the Premium Postdoctoral Fellowship Program of the Hungarian Academy of Sciences. HH is a Miguel Servet (CP14/00229) researcher funded by the Spanish Institute of Health Carlos III (ISCIII). This work has received funding from the Ministerio de Ciencia, Innovación y Universidades (SAF2017-89109-P; AEI/FEDER, UE)

Kernelized methods using multiple kernels have shown better performances in graph-based clustering. But those kernelized methods get affected by the noise present in the data set. Also, only a single view has been used in those kernelized graph-based clustering methods. To address those issues, a novel low-rank multi-view multi-kernel graph-based clustering framework (LRMVMKC) has been proposed in this paper. Where the similarity nature of kernel matrices are exploited by low-rank optimal kernel learning and the clustering performances are boosted by using multiple views that provide different partial information about a given data set. The use of the proposed LRMVMKC framework on different benchmark data sets demonstrates the better performances of the proposed framework over other existing methods.

Kernelized graph-based learning methods have gained popularity because of its better performances in the clustering task. But in high dimensional data, there exist many redundant features which may degrade the clustering performances. To solve this issue, we propose a novel multi-view kernelized graph-based clustering (MVKGC) framework for high dimensional data that performs the clustering task while reducing the dimensionality of the data. The proposed method also uses multiple views which help to improve the clustering performances by providing different partial information of a given data set. The extensive experiments of the proposed method on different real-world benchmark data sets show a better and efficient performance of the proposed method than other existing methods.