• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.4
• /
• pp.259-275
• /
• 2021

This study explored the usefulness and implications of the Bayesian hyperparameter optimization in developing species distribution models (SDMs). A variety of machine learning (ML) algorithms, namely, support vector machine (SVM), random forest (RF), boosted regression tree (BRT), XGBoost (XGB), and Multilayer perceptron (MLP) were used for predicting the occurrence of four benthic macroinvertebrate species. The Bayesian optimization method successfully tuned model hyperparameters, with all ML models resulting an area under the curve (AUC) > 0.7. Also, hyperparameter search ranges that generally clustered around the optimal values suggest the efficiency of the Bayesian optimization in finding optimal sets of hyperparameters. Tree based ensemble algorithms (BRT, RF, and XGB) tended to show higher performances than SVM and MLP. Important hyperparameters and optimal values differed by species and ML model, indicating the necessity of hyperparameter tuning for improving individual model performances. The optimization results demonstrate that for all macroinvertebrate species SVM and RF required fewer numbers of trials until obtaining optimal hyperparameter sets, leading to reduced computational cost compared to other ML algorithms. The results of this study suggest that the Bayesian optimization is an efficient method for hyperparameter optimization of machine learning algorithms.

• Journal of Electrochemical Science and Technology
• /
• v.9 no.4
• /
• pp.251-259
• /
• 2018

In general, the charge storage characteristics and overall performance of electrochemical energy devices (such as lithiumion batteries and supercapacitors) significantly depends on the structural and geometrical factors of the electrodes' active materials. The most widely used active materials of electrochemical energy storage devices are based on carbons of various forms. Each carbon type has drawbacks and advantages when used as the electrode material. Studies have been recently carried out to combine different types of carbons, in particular nanostructured carbons, in order to overcome the structure-originated limitations and thus enhance the overall electrochemical performances. In this feature article, we report the recent progress on the development of this novel class of materials (multidimensional nanocarbons), and their applications for supercapacitors. Multidimensional nanocarbons include graphenes/carbon nanotubes (CNTs), CNTs/carbon films, CNTs/fullerenes, and ternary carbon nanostructures. Various applications using these multidimensional nanocarbons have been proposed and demonstrated in the literature. Owing to the recent extensive studies on electrochemical energy storage devices and considering that carbons are their most fundamental electrode materials, the number of reports on nanocarbons employed as electrodes of the electrochemical energy storage devices is rapidly increasing. Recently, numerous multidimensional nanocarbons have been designed, prepared, and utilized as electrodes of electrochemical capacitors or supercapacitors, which are considered next-generation energy devices owing to their unique merits compared to the conventional structures. In this review, we summarize the basic motivations, preparation methods, and resultant supercapacitor performances of each class of multidimensional nanocarbons published in the literature, focusing on recent reports.

• STI Policy Review
• /
• v.9 no.1
• /
• pp.1-29
• /
• 2018

With the constant growth of R&D investment, it has been increasingly necessary to evaluate the effectiveness of R&D performance and there is a high emphasis on ensuring the accountability and effectiveness of R&D programs. The evaluation of performance of a firm is especially necessary in times of economic downturn to justify R&D investment. However, there is a marked shortage of clear guidelines as to where and how particular metrics are used to measure the output and outcome of R&D activity in firms. Many firms have difficulties in selecting appropriate indicators for their R&D and financial performances. To fill this gap, this article discusses and presents the findings from the literature in such a way that they become useful for researchers or managers who are in charge of measuring the R&D and business performances arising from innovation activities. Finally, based on the findings about metrics of R&D performance, this article proposes the hypothetical framework to investigate the relationship between technology forecasting, strategic technology planning, and business performance. The framework of this article will assist policy makers, universities, research institutes/national laboratories, and companies to enhance their decision making process in technology development.

• International Journal of Computer Science & Network Security
• /
• v.21 no.3
• /
• pp.41-47
• /
• 2021

Detection of protein-protein interactions (PPIs) remains essential for the development of therapies against diseases. Experimental studies to detect PPI are longer and more expensive. Today, with the availability of PPI data, several computer models for predicting PPIs have been proposed. One of the big challenges in this task is feature extraction. The relevance of the information extracted by some extraction techniques remains limited. In this work, we first propose an extraction method based on correlation relationships between the physicochemical properties of amino acids. The proposed method uses a correlation matrix obtained from the hydrophobicity and hydrophilicity properties that it then integrates in the calculation of the bigram. Then, we use the SVM algorithm to detect the presence of an interaction between 2 given proteins. Experimental results show that the proposed method obtains better performances compared to the approaches in the literature. It obtains performances of 94.75% in accuracy, 95.12% in precision and 96% in sensitivity on human HPRD protein data.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.1
• /
• pp.84-98
• /
• 2021

The hydrodynamic characteristics of amphibious assault vehicles are investigated using commercial CFD code, STAR-CCM+. Resistance performances of a displacement-type vehicle and a semi-planing type vehicle are analyzed in calm water. The self-propelled model is also computed for the semi-planing type vehicle. All computations are performed using an overset mesh system and a RANS based flow-solver coupled with a two-degree of freedom equations of motion. A moving reference frame is applied to simulate revolutions of impeller blades for a waterjet propulsion system. Grid dependency tests are performed to evaluate discretization errors for the mesh systems. The numerical analysis results are compared with the experimental results obtained from model tests. It is shown that RANS is capable of investigating the resistance and self-propulsion characteristics of high-speed amphibious assault vehicles. It is also found that a fully covered side skirt, which is covering tracks, reduces resistance and stern trim, besides increasing propulsive efficiency.

• Journal of Auto-vehicle Safety Association
• /
• v.14 no.3
• /
• pp.41-47
• /
• 2022

Composite material is recently very important material for eco-friendly vehicles because of its excellent mechanical property and lightweight effect. So, many research results have been recently published for developing the composite material to apply vehicles. In this paper, new brake system is presented using composite material to response this situation. And advantages in terms of performance compared to competitive company will be discussed in depth to verify superiorities of the new composite brake. To do so, composite brake systems which have the same size as the competitive company to the same vehicle is applied. And superiorities through a variety of test results are presented. First, normal braking performances are compared with competitive company through braking effect, heat capacity and friction test, Second, circuit driving and high speed fade test are also verified with competitive company to confirm harsh braking performances for the new composite brake system. Finally, the effects of applying the composite brake to automobile industry like electric car are analyzed.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.25 no.1
• /
• pp.115-132
• /
• 2023

The globalization of Korean culture has spurred a steady increase in the participation of domestic fashion designers in global performance art. This study analyzes the formal elements of stage costume designs by fashion designers, and further evaluates the impact of these designers' aesthetic philosophies on overall performance. This analysis seeks to provide a foundation for the development of new directionality in performance art, with an enhanced competitiveness derived from diversity and creativity. The scope of this research spans 15 performances and eight fashion designers' works in the post-2010s era. These performances are characterized by their break from traditional artistic norms, illustrating the modernization of the performing arts and reflecting the designers' aesthetic philosophies. This modernization incorporates inspirations drawn from a diverse selection of movements, such as fusion, culture clash, kitsch, and minimal avant-garde. As a result, in combination with a pure reflection of the designers' aesthetics, these designs heighten overall performance, suggest a new direction for traditional performance art, tap into a popular sensibility to expand communication to a wider audience, and push the boundaries of tradition through artistic innovation.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.20 no.4
• /
• pp.1-9
• /
• 2012

The NCF control algorithm for an active suspension system was proposed and investigated. The NCF algorithm using spring dynamic variation force and suspension relative velocity was applied to the 1/4 vehicle model and numerical analysis was performed. Vehicle's performances such as vehicle displacement, vehicle acceleration, suspension deflection, tire deflection and absorbed power were calculated and compared with those of the passive, semi-active and LQR active suspension system that use full state feedback. Numerical results show that the proposed NCF active suspension system has superior performance compared with the passive and semi-active suspension system and has very similar performance compared with the LQR active suspension system. So the proposed NCF algorithm is considered as a highly practical algorithm because it requires only one displacement sensor in a 1/4 vehicle model.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.1
• /
• pp.46-58
• /
• 2007

Current many software development projects gradually develop large volume of complex and changeable system. Architecture driven methodology is used for developing softwares like CBD (Component Based Development) or PLE (Product Line Engineering). To improve the development and maintainability in such an environment, the first of all needs the flexible and extensible architecture. This paper suggests the CBD application framework that has the flexbile and extensible architecture for improving development productivity and maintainability. In this paper, we also show basic performances of our suggested framework, comparing to a well-known web application framework, Struts.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.3
• /
• pp.143-147
• /
• 2014

The ECU which is operating inductive actuator like motor or solenoid should be designed with considering the heat removal performance and the EMC performance. In most cases, these two performances are trade-off. Especially, the metal cover with plastic housing for improving heat removal performance can affect EMC performance by coupling the noise source to harness cable and connector. Therefore, after analyzing the EMC effectiveness of the metal cover, countermeasures for EMC noise reduction should be established. In this paper, by simulating and testing of AHB Gen3 ECU, the influence of the metal cover to EMC performance is analyzed. And based on this result, we propose solutions for EMC noise reduction.