• Journal of Information Processing Systems
• /
• v.10 no.2
• /
• pp.314-333
• /
• 2014

When analyzing default predictions in real estate companies, the number of non-defaulted cases always greatly exceeds the defaulted ones, which creates the two-class imbalance problem. This lowers the ability of prediction models to distinguish the default sample. In order to avoid this sample selection bias and to improve the prediction model, this paper applies a minority sample generation approach to create new minority samples. The logistic regression, support vector machine (SVM) classification, and neural network (NN) classification use an imbalanced dataset. They were used as benchmarks with a single prediction model that used a balanced dataset corrected by the minority samples generation approach. Instead of using prediction-oriented tests and the overall accuracy, the true positive rate (TPR), the true negative rate (TNR), G-mean, and F-score are used to measure the performance of default prediction models for imbalanced dataset. In this paper, we describe an empirical experiment that used a sampling of 14 default and 315 non-default listed real estate companies in China and report that most results using single prediction models with a balanced dataset generated better results than an imbalanced dataset.

• Structural Engineering and Mechanics
• /
• v.73 no.5
• /
• pp.565-584
• /
• 2020

The nonlinear thermo-electro-elastic buckling behavior of viscoelastic nanoplates under magnetic field is investigated based on nonlocal elasticity theory. Employing nonlinear strain-displacement relations, the geometrical nonlinearity is modeled while governing equations are derived through Hamilton's principle and they are solved applying semi-analytical generalized differential quadrature (GDQ) method. Eringen's nonlocal elasticity theory considers the effect of small size, which enables the present model to become effective in the analysis and design of nano-sensors and nano actuators. Based on Kelvin-Voigt model, the influence of the viscoelastic coefficient is also discussed. It is demonstrated that the GDQ method has high precision and computational efficiency in the buckling analysis of viscoelastic nanoplates. The good agreement between the results of this article and those available in literature validated the presented approach. The detailed mathematical derivations are presented and numerical investigations are performed while the emphasis is placed on investigating the effect of the several parameters such as electric voltage, small scale effects, elastomeric medium, magnetic field, temperature effects, the viscidity and aspect ratio of the nanoplate on its nonlinear buckling characteristics. It is explicitly shown that the thermo-electro-elastic nonlinear buckling behavior of viscoelastic nanoplates is significantly influenced by these effects. Numerical results are presented to serve as benchmarks for future analyses of viscoelastic nanoplates as fundamental elements in nanoelectromechanical systems.

• Journal of Information Technology Services
• /
• v.18 no.3
• /
• pp.143-154
• /
• 2019

The purpose of this study is to explore innovative adaptation of IT systems worldwide to support process-focus assessment. To this end, the study presents four cases from Australia, Singapore, the U.S.A., and Korea to inform educational policy and technology researchers and practitioners. Based on comparing the four chosen cases as benchmarks, the study concluded that IT systems and technologies can expedite and improve school interventions to enhance student learning in terms of time and quality. Also, educational big data and learning analytics can be used to systematically monitor and communicate individual student's progress among school stakeholders (i.e., teachers, students, parents, and administrators). Lastly, the study made some suggestions to support process-focused assessment in schools as following : 1) A more evidence-based, systems approach is needed to integrate the curriculum, instruction, and assessment to bridge the gap between educational policy and school practice; 2) It is critical to create ICT-friendly school environments so that meaningful data could be collected, analyzed, and stored from individual students and school units; 3) Teacher professional development is another area that needs special considerations and support to successfully implement process-focused assessment in schools.

• Advances in nano research
• /
• v.4 no.2
• /
• pp.85-111
• /
• 2016

In this paper, the classical and non-classical boundary conditions effect on free vibration characteristics of functionally graded (FG) size-dependent nanobeams are investigated by presenting a semi analytical differential transform method (DTM) for the first time. Three kinds of mathematical models, namely; power law (P-FGM), sigmoid (S-FGM) and Mori-Tanaka (MT-FGM) distribution are considered to describe the material properties in the thickness direction. The nonlocal Eringen theory takes into account the effect of small size, which enables the present model to become effective in the analysis and design of nanosensors and nanoactuators. Governing equations are derived through Hamilton's principle and they are solved applying semi analytical differential transform method. The good agreement between the results of this article and those available in literature validated the presented approach. The detailed mathematical derivations are presented and numerical investigations are performed while the emphasis is placed on investigating the effect of the several parameters such as small scale effects, spring constant factors, various material compositions and mode number on the normalized natural frequencies of the FG nanobeams in detail. It is explicitly shown that the vibration of FG nanobeams is significantly influenced by these effects. Numerical results are presented to serve as benchmarks for future analyses of FG nanobeams.

• ETRI Journal
• /
• v.43 no.3
• /
• pp.524-537
• /
• 2021

High-performance Linpack (HPL) is among the most popular benchmarks for evaluating the capabilities of computing systems and has been used as a standard to compare the performance of computing systems since the early 1980s. In the initial system-design stage, it is critical to estimate the capabilities of a system quickly and accurately. However, the original HPL mathematical model based on a single core and single communication layer yields varying accuracy for modern processors and accelerators comprising large numbers of cores. To reduce the performance-estimation gap between the HPL model and an actual system, we propose a mathematical model for multi-communication layered HPL. The effectiveness of the proposed model is evaluated by applying it to a GPU cluster and well-known systems. The results reveal performance differences of 1.1% on a single GPU. The GPU cluster and well-known large system show 5.5% and 4.1% differences on average, respectively. Compared to the original HPL model, the proposed multi-communication layered HPL model provides performance estimates within a few seconds and a smaller error range from the processor/accelerator level to the large system level.

• Steel and Composite Structures
• /
• v.30 no.6
• /
• pp.493-516
• /
• 2019

This paper is dedicated to nonlinear static and free vibration analysis of Uniform Distributed Carbon Nanotube Reinforced Composite (UD-CNTRC) structures under in-plane loading. The authors have suggested an efficient six-node triangular element. Mixed Interpolation of Tensorial Components (MITC) approach is employed to alleviate the membrane locking phenomena. Moreover, the behavior of the well-known LST element is considerably improved by applying an additional linear interpolation on the strain fields. Based on the rule of mixture, the properties of CNTRC are obtained. In this study, only the uniform distributed CNTs are employed through the thickness direction of element. To achieve the natural frequencies and shape modes, the eigenvalue problem is also solved. Using Total Lagrangian Principles, large amplitude free vibration is considered based on the first normalized mode shape of structure. Different well-known plane problem benchmarks and some proposed ones are studied to validate the accuracy and capability of authors' formulations. In addition, the effects of length to the height ratio of beam, CNT's characteristics, support conditions and normalized amplitude parameter on the linear and nonlinear vibration parameters are investigated.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.3182-3198
• /
• 2019

Vision-based Human Pose Estimation has been considered as one of challenging research subjects due to problems including confounding background clutter, diversity of human appearances and illumination changes in scenes. To tackle these problems, we propose to use a new multi-stage convolution machine for estimating human pose. To provide better heatmap prediction of body joints, the proposed machine repeatedly produces multiple predictions according to stages with receptive field large enough for learning the long-range spatial relationship. And stages are composed of various modules according to their strategic purposes. Pyramid stacking module and dilation module are used to handle problem of human pose at multiple scales. Their multi-scale information from different receptive fields are fused with concatenation, which can catch more contextual information from different features. And spatial and channel information of a given input are converted to gating factors by squeezing the feature maps to a single numeric value based on its importance in order to give each of the network channels different weights. Compared with other ConvNet-based architectures, we demonstrated that our proposed architecture achieved higher accuracy on experiments using standard benchmarks of LSP and MPII pose datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.1
• /
• pp.86-109
• /
• 2019

Cloud computing systems' performance is still a central focus of research for determining optimal resource utilization. Running several existing benchmarks simultaneously serves to acquire performance information from specific cloud system resources. However, the complexity of monitoring the existing performance of computing systems is a challenge requiring an efficient and interactive user directing performance-monitoring system. In this paper, we propose an effective multivariate control framework for monitoring cloud systems performance. The proposed framework utilizes the hardware cloud systems performance metrics, collects and displays the performance measurements in terms of meaningful graphics, stores the graphical information in a database, and provides the data on-demand without requiring a third party software. We present performance metrics in terms of CPU usage, RAM availability, number of cloud active machines, and number of running processes on the selected machines that can be monitored at a high control level by either using a cloud service customer or a cloud service provider. The experimental results show that the proposed framework is reliable, scalable, precise, and thus outperforming its counterparts in the field of monitoring cloud performance.

• Genomics & Informatics
• /
• v.17 no.2
• /
• pp.17.1-17.3
• /
• 2019

Text mining has become an important research method in biology, with its original purpose to extract biological entities, such as genes, proteins and phenotypic traits, to extend knowledge from scientific papers. However, few thorough studies on text mining and application development, for plant molecular biology data, have been performed, especially for rice, resulting in a lack of datasets available to solve named-entity recognition tasks for this species. Since there are rare benchmarks available for rice, we faced various difficulties in exploiting advanced machine learning methods for accurate analysis of the rice literature. To evaluate several approaches to automatically extract information from gene/protein entities, we built a new dataset for rice as a benchmark. This dataset is composed of a set of titles and abstracts, extracted from scientific papers focusing on the rice species, and is downloaded from PubMed. During the 5th Biomedical Linked Annotation Hackathon, a portion of the dataset was uploaded to PubAnnotation for sharing. Our ultimate goal is to offer a shared task of rice gene/protein name recognition through the BioNLP Open Shared Tasks framework using the dataset, to facilitate an open comparison and evaluation of different approaches to the task.

• Structural Engineering and Mechanics
• /
• v.72 no.3
• /
• pp.329-338
• /
• 2019

In this paper, the magnetic field influence on the wave propagation characteristics of graphene nanosheets is examined within the frame work of a two-variable plate theory. The small-scale effect is taken into consideration based on the nonlocal strain gradient theory. For more accurate analysis of graphene sheets, the proposed theory contains two scale parameters related to the nonlocal and strain gradient effects. A derivation of the differential equation is conducted, employing extended principle of Hamilton and solved my means of analytical solution. A refined trigonometric two-variable plate theory is employed in Kinematic relations. The scattering relation of wave propagation in solid bodies which captures the relation of wave number and the resultant frequency is also investigated. According to the numerical results, it is revealed that the proposed modeling can provide accurate wave dispersion results of the graphene nanosheets as compared to some cases in the literature. It is shown that the wave dispersion characteristics of graphene sheets are influenced by magnetic field, elastic foundation and nonlocal parameters. Numerical results are presented to serve as benchmarks for future analyses of graphene nanosheets.