Korean automatic spacing using pretrained transformer encoder and analysis

Automatic spacing in Korean is used to correct spacing units in a given input sentence. The demand for automatic spacing has been increasing owing to frequent incorrect spacing in recent media, such as the Internet and mobile networks. Therefore, herein, we propose a transformer encoder that reads a sentence bidirectionally and can be pretrained using an out-of-task corpus. Notably, our model exhibited the highest character accuracy (98.42%) among the existing automatic spacing models for Korean. We experimentally validated the effectiveness of bidirectional encoding and pretraining for automatic spacing in Korean. Moreover, we conclude that pretraining is more important than fine-tuning and data size.     

Automatic logo transition detection in digital video contents

The amount of user created contents has been increasing rapidly and is associated with a serious copyright problem. Automatic logo detection and recognition in videos is a natural and efficient way of overcoming the copyright problem. However, logos have varying characteristics, which make logo detection and recognition very difficult. Moreover, logo transitions between two different logos exist in one video comprising several video contents. This disrupts the automatic logo detection and recognition. Therefore, in order to improve logo detection, it is necessary to take into account the logo transitions explicitly. This paper proposes an accurate logo transition detection method for recognizing logos in digital video contents. The proposed method accurately segments a video according to logo and efficiently recognizes various types of logos. The experimental results demonstrate the effectiveness of the proposed method for logo detection and video segmentation according to logo.     

A numerical study on the performance evaluation of ventilation systems for indoor radon reduction

Numerical simulations were conducted using computational fluid dynamics to evaluate the effect of ventilation conditions on radon (²²²Rn) reduction performance in a residential building. The results indicate that at the same ventilation rate, a mechanical ventilation system is more effective in reducing indoor radon than a natural ventilation system. For the same ventilation type, the indoor radon concentration decreases as the ventilation rate increases. When the air change per hour (ACH) was 1, the indoor radon concentration was maintained at less than 100 Bq/m³. However, when the ACH was lowered to 0.01, the average indoor radon concentration in several rooms exceeded 148 Bq/m³. The angle of the inflow air was found to affect the indoor air stream and consequently the distribution of the radon concentration. Even when the ACH was 1, the radon concentrations of some areas were higher than 100 Bq/m³ for inflow air angles of 5° and 175°.     

Synthesis of mesoporous silica in acidic condition by solvent evaporation method

Both powder-type and film-type mesoporous silica materials have been prepared by using a sol-gel process in highly acidic, solvent-rich and mild conditions by solvent evaporation method. Powder mesoporous silica which has hexagonal arrays was well synthesized within only 10 minutes, and mesoporous film was formed on the glass plate within only several seconds by supramolecular interactions between the silicate species and surfactant-templates. In particular, the synthesis time was very short compared with that of hydrothermal synthesis owing to the solvent evaporation. This method has some characteristics different from the hydrothermal method such as the following. There is no precipitation because particles do not form in the solution before solvent evaporation, bulk powder is rigid, whereas that of hydrothermal synthesis is very soft, and mesoporous materials are synthesized in acidic condition. Solvent evaporation method for the synthesis of mesoporous materials is promising because the method is very simple and time-saving.     

High‐Performance Sodium‐Ion Hybrid Supercapacitor Based on Nb2O5@Carbon Core–Shell Nanoparticles and Reduced Graphene Oxide Nanocomposites

Sodium‐ion hybrid supercapacitors (Na‐HSCs) have potential for mid‐ to large‐scale energy storage applications because of their high energy/power densities, long cycle life, and the low cost of sodium. However, one of the obstacles to developing Na‐HSCs is the imbalance of kinetics from different charge storage mechanisms between the sluggish faradaic anode and the rapid non‐faradaic capacitive cathode. Thus, to develop high‐power Na‐HSC anode materials, this paper presents the facile synthesis of nanocomposites comprising Nb₂O₅@Carbon core–shell nanoparticles (Nb₂O₅@C NPs) and reduced graphene oxide (rGO), and an analysis of their electrochemical performance with respect to various weight ratios of Nb₂O₅@C NPs to rGO (e.g., Nb₂O₅@C, Nb₂O₅@C/rGO‐70, ‐50, and ‐30). In a Na half‐cell configuration, the Nb₂O₅@C/rGO‐50 shows highly reversible capacity of ≈285 mA h g^?1 at 0.025 A g^?1 in the potential range of 0.01–3.0 V (vs Na/Na⁺). In addition, the Na‐HSC using the Nb₂O₅@C/rGO‐50 anode and activated carbon (MSP‐20) cathode delivers high energy/power densities (≈76 W h kg^?1 and ≈20 800 W kg^?1) with a stable cycle life in the potential range of 1.0–4.3 V. The energy and power densities of the Na‐HSC developed in this study are higher than those of similar Li‐ and Na‐HSCs previously reported.     

Defect diagnostics of gas turbine engine using hybrid SVM-ANN with module system in off-design condition

A hybrid method of an artificial neural network (ANN) and a support vector machine (SVM) has been used for a health monitoring algorithm of a gas turbine engine. The method has the advantage of reducing learning data and converging time without any loss of estimation accuracy, because the SVM classifies the defect location and reduces the learning data range. In off-design condition, however, the operation region of the engine becomes wide and the nonlinearity of learning data increases considerably. Therefore, an improved hybrid method with the module system and the advanced SVM has been suggested to solve the problems. The module system divides the whole operating region into reasonably small-sized sections, and the advanced SVM has two steps of the classification. The proposed algorithm has been proven to reliably and effectively diagnose the simultaneous defects of the triple components as well as the defects of the single and dual components of the gas turbine engine in off-design condition. This paper was recommended for publication in revised form by Associate Editor Tong Seop Kim Tae-Seong Roh received his B.S. and M.S. degrees in Aeronautical Engineering from Seoul National University in 1984 and 1986. He then went on to receive his Ph.D. degree from Pennsylvania State University in 1995. Dr. Roh is currently a Professor at the department of Aerospace Engi-neering at Inha University in Incheon, Korea. His research interests are in the area of combustion instabilities, rocket and jet propulsions, interior ballistics, and gas turbine engine defect diagnostics. Dong-Whan Choi received his B.S. degree in Aeronautical Engineering from Seoul National University in 1974. He then went on to receive his M.S. and Ph.D. degrees from University of Washington in 1978 and 1983. Dr. Choi served three years as a President of Korea Aerospace Research Institute from 1999. He is currently a professor at the department of Aerospace Engineering at Inha University in Incheon, Korea. His research interests are in the area of turbulence, jet propulsions, and gas turbine defect diagnostics.     

Backward-motion control of a mobile robot with n passive off-hooked trailers

A passive multiple-trailer system provides various practical advantages for multi-functional service robots. However, motion control is difficult because the kinematic model is highly nonlinear. The kinematic design of a trailer system was proposed in prior research of ours. In this paper, it is shown how the backward motion of a robot with n passive trailers can be controlled. Once the desired trajectory of the last trailer is computed, the control input of the pushing robot is obtained through the proposed control scheme. Some experimental issues on reversing the trailer system are addressed. This paper provides an answer to the following question: “Does the system work well even if there are sensing or modeling errors?” Although it is difficult to obtain general analytic solutions for the above research question, a practical answer will be explored though simplified analysis and experiments. Experimental verifications are carried out using a mobile robot with three passive trailers. The experimental results show that backward-motion control can be successfully carried out by applying the proposed control scheme.     

Mechanism of droplet generation in silver thin films for organic light-emitting diode displays

Thin silver film is widely used as the cathode in organic light-emitting diode displays and it is generally fabricated using the thermal evaporation method. But during the evaporation process, there is an inevitable outgassing problem and this creates high viscosity bubbles in melted silver. When the bubbles break, the high energy released scatters silver droplets which damage the silver surface. In this study, we were able to decrease the number of droplets from 6,171 to 278 with a degassing process of 400 °C for 6 h before proceeding with a thermal evaporation process.     

Advancement of MIMO technology in WiMAX: from IEEE 802.16d/e/j to 802.16m

WiMAX is the first cellular standard that employs OFDMA technology and provides true integrated services for both fixed and mobile broadband access. Among the many new technologies adopted in WiMAX, MIMO antenna technology plays an essential role in delivering fast, rich-content, mobile broadband service reliably over extended coverage areas. In this article we provide a survey on the state of art of MIMO technologies in current WiMAX standards with an emphasis on practical engineering considerations. Moreover, we also briefly discuss the ongoing MIMO technologies in the evolution toward the next-generation WiMAX network.