• 대한음성학회지:말소리
• /
• 제65호
• /
• pp.105-124
• /
• 2008

This paper presents the effects of different phone-like-unit (PLU) sets in order to propose an optimal PLU set for the performance improvement of Korean automatic speech recognition (ASR) systems. The examination of 9 currently used PLU sets indicates that most of them include a selection of allophones without any sufficient phonetic base. In this paper, a total of 34 PLU sets are designed based on Korean phonetic characteristics arid the effects of each PLU set are evaluated through experiments. The results show that the accuracy rate of each phone is influenced by different phonetic constraint(s) which determine(s) the PLU sets, and that an optimal PLU set can be anticipated through the phonetic analysis of the given speech data.

• 한국전산유체공학회지
• /
• 제6권3호
• /
• pp.19-26
• /
• 2001

An Eulerian-Lagrangian method, so called immersed boundary method, is used for analysing viscous flow around arbitrary bodies, where governing equations are discretized on a regular grid by using a finite volume method. To improve the accuracy of flow near body boundaries, a second-order accurate interpolation scheme is used and a level-set based grid deformation method is presented to construct the adaptive grids around body boundaries. The present scheme is used to simulate steady flow around a semicircular cylinder mounted on the bottom of flow domain and calculated results are validated by results of a body fitted grid method. Finally, present method is applied to a complex flow around multi body and the usefulness is checked by investigating calculated results.

• Structural Engineering and Mechanics
• /
• 제87권3호
• /
• pp.243-254
• /
• 2023

Engineering structures usually suffer from cracks. The crack geometry has an influence on the structural mechanical properties and subsequent crack propagations. However, as an extensively utilized method in fracture analysis, the extended finite element method provided by Abaqus fails to output the specific location and dimensions of fractures. In this study, a technique to capture the crack geometry is proposed. The technique is based on the invariant level set method (I-LSM), which can avoid updating the level set function during crack development. The solution is achieved by an open-source plug-in programmed by Python. Three examples were performed to verify the effectiveness and robustness of the program. The result shows that the developed program can accurately output the crack geometry in both the 2D and 3D models. The open-source plug-in codes are included as supplementary material.

• 한국IT서비스학회지
• /
• 제20권2호
• /
• pp.13-31
• /
• 2021

Recently, delivery app services based on the O2O platform are increasing rapidly. Accordingly, various studies on O2O service have been conducted. Most of the studies are on consumer behavior in O2O services, and few studies on platform vendors have been conducted. Therefore, this study empirically analyzed the factors affecting the vender's intention to continuous use in the O2O delivery app platform service. Based on prior researches, we set the quality characteristics and network characteristics of the O2O platform as independent variables. The quality characteristics of the O2O platform consisted of system quality, information quality, and service quality, and the O2O platform network characteristics consisted of network externality and platform reputation. Perceived value and switching cost were set as mediated variables, and vender's intention to continuous use was set as dependent variables. For empirical analysis, we conducted a survey targeting vendors of O2O delivery app platform service, and conducted frequency analysis, factor analysis, reliability analysis, and regression analysis. As a result of the analysis, the quality characteristics of the O2O platform, such as system quality, information quality, service quality, and O2O platform network characteristics, showed that network externality and platform reputation had a positive effect on perceived value. The perceived value was found to have a positive effect on the switching cost and the intention to continuous use, and the switching cost was found to mediate the perceived value and the intention to continuous use. This study can contribute to the establishment of platform operation strategy as an empirical analysis on the factors that influence the intention of O2O platform vendors to use the platform continuously.

• 한국비블리아학회지
• /
• 제25권3호
• /
• pp.101-118
• /
• 2014

컴퓨터 정보기술의 발전과 함께 감정 기반 컴퓨팅이 다양한 분야에 빠르게 발전하여 확산되고 있다. 감정 기반 컴퓨팅의 지속적인 발전을 위해서는 이미지와 같은 멀티미디어의 콘텐츠의 감정 기반 색인과 검색이 필수적이다. 그러나 감정과 같은 추상적 개념은 주관적이며 이미지의 하위 수준 속성에서 유추하는데 한계가 있기 때문에 감정 색인은 통상적으로 난제로 여겨진다. 본 연구는 감정 색인 개선을 위해서 이미지에 대하여 이용자가 느끼는 감정 반응을 활용하여 이미지를 감정으로 접근하는데 있어서 확장된 접근점을 제공하는 방안을 고찰하였다. 이를 위하여 유로피아나 DB에서 사랑, 행복, 분노, 공포, 슬픔의 5가지 기본 감정을 표현한 이미지 15건을 선정하여 20명의 연구 참여자에게 보여주고 용어를 수집하였다. 이용자의 이미지 반응에서 수집한 용어는 정련 후 총 399건의 고유한 용어로 나타났다. 고유한 399건의 용어는 전체 1,093회 출현하였으며, 동시출현단어분석을 수행하여 상위 출현한 용어 네트워크를 구현하였다. 동시출현단어분석 기반의 네트워크를 통해서 기본 감정 용어와 함께 빈번하게 출현하는 용어를 규명하였다. 이를 통해 기본 감정용어와 함께 확장되어 제시될 수 있는 용어는 형용사, 동작/행위 표현 등 다양하게 나타났다.

• 산업경영시스템학회지
• /
• 제45권4호
• /
• pp.157-166
• /
• 2022

In the case of a die-casting process, defects that are difficult to confirm by visual inspection, such as shrinkage bubbles, may occur due to an error in maintaining a vacuum state. Since these casting defects are discovered during post-processing operations such as heat treatment or finishing work, they cannot be taken in advance at the casting time, which can cause a large number of defects. In this study, we propose an approach that can predict the occurrence of casting defects by defect type using machine learning technology based on casting parameter data collected from equipment in the die casting process in real time. Die-casting parameter data can basically be collected through the casting equipment controller. In order to perform classification analysis for predicting defects by defect type, labeling of casting parameters must be performed. In this study, first, the defective data set is separated by performing the primary clustering based on the total defect rate obtained during the post-processing. Second, the secondary cluster analysis is performed using the defect rate by type for the separated defect data set, and the labeling task is performed by defect type using the cluster analysis result. Finally, a classification learning model is created by collecting the entire labeled data set, and a real-time monitoring system for defect prediction using LabView and Python was implemented. When a defect is predicted, notification is performed so that the operator can cope with it, such as displaying on the monitoring screen and alarm notification.

• 한국전문물리치료학회지
• /
• 제27권2호
• /
• pp.140-148
• /
• 2020

Background: Scapular winging (SW) could be caused by tightness or weakness of the periscapular muscles. Although data mining techniques are useful in classifying or predicting risk of musculoskeletal disorder, predictive models for risk of musculoskeletal disorder using the results of clinical test or quantitative data are scarce. Objects: This study aimed to (1) investigate the difference between young women with and without SW, (2) establish a predictive model for presence of SW, and (3) determine the cutoff value of each variable for predicting the risk of SW using the decision tree method. Methods: Fifty young female subjects participated in this study. To classify the presence of SW as the outcome variable, scapular protractor strength, elbow flexor strength, shoulder internal rotation, and whether the scapula is in the dominant or nondominant side were determined. Results: The classification tree selected scapular protractor strength, shoulder internal rotation range of motion, and whether the scapula is in the dominant or nondominant side as predictor variables. The classification tree model correctly classified 78.79% (p = 0.02) of the training data set. The accuracy obtained by the classification tree on the test data set was 82.35% (p = 0.04). Conclusion: The classification tree showed acceptable accuracy (82.35%) and high specificity (95.65%) but low sensitivity (54.55%). Based on the predictive model in this study, we suggested that 20% of body weight in scapular protractor strength is a meaningful cutoff value for presence of SW.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2007년도 춘계 학술대회논문집
• /
• pp.95-98
• /
• 2007

A general purpose program NUFLEX has been extended for two-phase flows with topologically complex interface and cavitation flows with liquid-vapor phase change caused by large pressure drop. In analysis of two-phase flow, the phase interfaces are tracked by employing a LS(Level Set) method. Compared with the VOF(Volume-of-Fluid} method based on a non-smooth volume-fraction function, the LS method can calculate an interfacial curvature more accurately by using a smooth distance function. Also, it is quite straightforward to implement for 3-D irregular meshes compared with the VOF method requiring much more complicated geometric calculations. Also, the cavitation process is computed by including the effects of evaporation and condensation for bubble formation and collapse as well as turbulence in flows. The volume-faction and continuity equations are adapted for cavitation models with phase change. The LS and cavitation formulation are implemented into a general purpose program for 3-D flows and verified through several test problems.

• 설비공학논문집
• /
• 제19권5호
• /
• pp.411-417
• /
• 2007

This study has been conducted to investigate pumping characteristics of diffuser/nozzle based piezoelectric micropumps. The micropumps include a piezo disk (an actuator), a chamber and a set of diffuser and nozzle. Flow in the current micropumps is controlled by a set of diffuser and nozzle, not by a nap valve. The diffuser/nozzle based micropumps are more reliable in operation and are easier in manufacturing than the flap valve based micropumps. The flow rates of the piezoelectric micropumps have been closely analyzed with a numerical calculation. It has been found that the positions of the inlet and outlet of the micropump can influence the performance of the diffuser/nozzle based piezoelectric micropumps. This study may provide fundamental understanding for the design and analysis of the piezoelectric micropumps.

• Design & Manufacturing
• /
• 제16권3호
• /
• pp.34-38
• /
• 2022

Regular expression pattern matching is widely used in applications such as computer virus vaccine, NIDS and DNA sequencing analysis. Hardware-based pattern matching is used when high-performance processing is required due to time constraints. ReCPU, SMPU, and REMP, which are processor-based regular expression matching processors, have been proposed to solve the problem of the hardware-based method that requires resynthesis whenever a pattern is updated. However, these processor-based regular expression matching processors inefficiently handle repetitive operations of regular expressions. In this paper, we propose a new instruction set to improve the inefficient repetitive operations of ReCPU and SMPU. We propose REMPi, a regular expression matching processor that enables efficient iterative operations based on the REMP instruction set. REMPi improves the inefficient method of processing a particularly short sub-pattern as a repeat operation OR, and enables processing with a single instruction. In addition, by using a down counter and a counter stack, nested iterative operations are also efficiently processed. REMPi was described with Verilog and synthesized on Intel Stratix IV FPGA.