• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.45 no.4
• /
• 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.

• Annual Conference on Human and Language Technology
• /
• 1998.10c
• /
• pp.379-385
• /
• 1998

한국어가 연속적으로 발음될 때 여러 가지 음운 변동현상이 일어난다. 이것은 한국어 연속음성 인식을 어렵게 하는 주요 요인 중의 한가지이다. 본 논문은 음운변동현상이 반영된 음성 인식 문자열을 규칙에 의거하여 text 기반 문자열로 다시 복원시키고 복원 결과 후보들을 형태소 분석하여 유용한 문자열만을 최종 결과로 생성하게 하는 시스템을 구성하였다. 복원은 4가지 rule 즉, 음절 경계 종성 초성 복원 rule, 모음처리 복원 rule, 끝음절 중성 복원 rule, 한 음절처리 rule에 따라 이루어진다. 규칙 적용 과정중에 효과적인 복원을 위해 x-clustering정보를 정의 하여 사용하고, 형태소 분석기에 입력될 복원 후보수를 제한하기 위해 postfix음절 빈도정보를 구하여 사용한다.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2006.08a
• /
• pp.1219-1224
• /
• 2006

This study investigates the effect of $NH_3$ gas upon the growth of carbon nanotubes (CNTs) using thermal chemical vapor deposition. It is considered that the CNT synthesis occurs mainly through two steps, clustering of catalyst particles and subsequent growth of CNTs. We thus introduced $NH_3$ during either an annealing or growth step. When $NH_3$ was fed only during annealing, CNTs grew longer and more highly crystalline with diameters unchanged. An addition of $NH_3$ during growth, however, resulted in shorter CNTs with lower crystallinity while increased their diameters. Vertically aligned, highly populated CNT samples showed poor field emission characteristics, leading us to apply post-treatments onto the CNT surface. The CNTs were treated by adhesive tapes or etched back by dc plasma of $N_2$ to reduce the population density and the radius of curvatures of CNTs. We discuss the morphological changes of CNTs and their field emission properties upon surface treatments.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2004.04b
• /
• pp.30-33
• /
• 2004

Shallow p+-n junctions were formed by low-energy ion implantation and dual-step annealing processes The dopant implantation was performed into the crystalline substrates using $BF_2$ ions. The annealing was performed with a rapid thermal processor and a furnace. FA+RTA annealing sequence exhibited better junction characteristics than RTA+FA thermal cycle from the viewpoint of junction depth. A new simulator is designed to model boron diffusion in silicon, which is especially useful for analyzing the annealing process subsequent to ion implantation. The model which is used in this simulator takes into account nonequilibrium diffusion, reactions of point defects, and defect-dopant pairs considering their charge states, and the dopant inactivation by introducing a boron clustering reaction. Using a resonable parameter values, the simulator covers not only the equilibrium diffusion conditions but also the nonequilibrium post-implantation diffusion. Using initial conditions and boundary conditions, coupled diffusion equation is solved successfully. The simulator reproduced experimental data successfully.

• The Journal of Korean Medicine
• /
• v.25 no.1
• /
• pp.172-187
• /
• 2004

Objectives : To characterize four types of obesity and to effectively improve the treatment of obesity through Oriental medicine Methods : At 00 Oriental Medical Center, 203 female subjects who intented to lose weight were requested to complete a questionnaire. These participants were also given physical tests. The Questionnaire consisted of questions both about general life style and obesity in oriental medicine framework. The physical tests were blood tests, a body composition via Inbody 2.0, and physical strength tests. One-way analysis of variance was done to compare the means of items and physical tests between four types of obesity. Duncan was used by post hoc test. Results : 1. Significant differences between obesity type III and obesity type IV in ever tried to lose weight, childhood obesity and excercise times were observed in the questionnaire of general life style(p<0.05). 2. Ducan test showed significant differences between four obesity types in diseases (p<0.05). 3. Significant differences between four obesity types in height, % body fat muscular endurance, soft lean mass, fat mass, Trigliceride, Total cholesterol and ${\gamma}-GTP$ were observed (p<0.05). Conclusions : Further clinical research is necessary in the four types of obesity explored. The diagnosis and treatment based on these types should be further studied.

• Atmosphere
• /
• v.21 no.3
• /
• pp.221-228
• /
• 2011

The nonlinear characteristics of summer monsoon intraseasonal oscillation (ISO) in precipitation, which is manifested as fluctuations in convection and circulation, is one of the major difficulty on the prediction of East Asian summer monsoon (EASM). The present study aims to identify the spatial distribution and time evolution of nonlinear phases of monsoon ISO. In order to classify the different phases of monsoon ISO, Self-Organizing Map(SOM) known as a nonlinear pattern recognition technique is used. SOM has a great attractiveness detecting self-similarity among data elements by grouping and clustering such self-similar components. The four important patterns are demonstrated as Meiyu-Baiu, Changma, post-Changma, and dry-spell modes. It is found that SOM well captured the formation of East Asian monsoon trough during early summer and its northward migration together with enhanced convection over subtropical western Pacific and regionally intensive precipitation including Meiyu, Changma and Baiu. The classification of fundamental large scale spatial pattern and evolutionary history of nonlinear phases of monsoon ISO provides the source of predictability for extended-range forecast of summer precipitation.

• Journal of Microbiology and Biotechnology
• /
• v.24 no.11
• /
• pp.1473-1483
• /
• 2014

Reports of herbicide resistance events are proliferating worldwide, leading to new cultivation strategies using combinations of pre-emergence and post-emergence herbicides. We analyzed the impact during a one-year cultivation cycle of several herbicide combinations on the rhizobacterial community of glyphosate-tolerant Bt-maize and compared them to those of the untreated or glyphosate-treated soils. Samples were analyzed using pyrosequencing of the V6 hypervariable region of the 16S rRNA gene. The sequences obtained were subjected to taxonomic, taxonomy-independent, and phylogeny-based diversity studies, followed by a statistical analysis using principal components analysis and hierarchical clustering with jackknife statistical validation. The resilience of the microbial communities was analyzed by comparing their relative composition at the end of the cultivation cycle. The bacterial communites from soil subjected to a combined treatment with mesotrione plus s-metolachlor followed by glyphosate were not statistically different from those treated with glyphosate or the untreated ones. The use of acetochlor plus terbuthylazine followed by glyphosate, and the use of aclonifen plus isoxaflutole followed by mesotrione clearly affected the resilience of their corresponding bacterial communities. The treatment with pethoxamid followed by glyphosate resulted in an intermediate effect. The use of glyphosate alone seems to be the less aggressive one for bacterial communities. Should a combined treatment be needed, the combination of mesotrione and s-metolachlor shows the next best final resilience. Our results show the relevance of comparative rhizobacterial community studies when novel combined herbicide treatments are deemed necessary to control weed growth.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.5
• /
• pp.637-644
• /
• 2006

Biomedical image is often complex. An applied image analysis system should deal with the images which are of quite low quality and are challenging to segment. This paper presents a framework for color cell image segmentation by learning and classification online. It is a robust two-stage scheme using kernel method and watershed transform. In first stage, a two-class SVM is employed to discriminate the pixels of object from background; where the SVM is trained on the data which has been analyzed using the mean shift procedure. A real-time training strategy is also developed for SVM. In second stage, as the post-processing, local watershed transform is used to separate clustering cells. Comparison with the SSF (Scale space filter) and classical watershed-based algorithm (those are often employed for cell image segmentation) is given. Experimental results demonstrate that the new method is more accurate and robust than compared methods.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1480-1493
• /
• 2018

Recently, computational intelligence has received a lot of attention from researchers due to its potential applications to artificial intelligence. In computer science, computational intelligence refers to a machine's ability to learn how to compete various tasks, such as making observations or carrying out experiments. We adopted a computational intelligence solution to monitoring residual resources in cloud computing environments. The proposed residual resource monitoring scheme periodically monitors the cloud-based host machines, so that the post migration performance of a virtual machine is as consistent with the pre-migration performance as possible. To this end, we use a novel similarity measure to find the best target host to migrate a virtual machine to. The design of the proposed residual resource monitoring scheme helps maintain the quality of service and service level agreement during the migration. We carried out a number of experimental evaluations to demonstrate the effectiveness of the proposed residual resource monitoring scheme. Our results show that the proposed scheme intelligently measures the similarities between virtual machines in cloud computing environments without causing performance degradation, whilst preserving the quality of service and service level agreement.

• International Journal of Advanced Culture Technology
• /
• v.10 no.1
• /
• pp.204-210
• /
• 2022

From the 1970s to the present, the number of new infectious diseases such as SARS, Ebola virus, and MERS has steadily increased. The new infectious disease, COVID-19, which began in Wuhan, Hubei Province, China, has pushed the world into a pandemic era. As a result, Countries imposed restrictions on entry to foreign countries due to concerns over the spread of COVID-19, which led to a decrease in the movement of tourists. Due to the restriction of travel, keywords such as "Corona blue" have soared and depression has increased. Therefore, this study aims to analyze the stress meaning network of the COVID-19 era to derive keywords and come up with a plan for a travel-related platform of the Post-COVID 19 era. This study conducted analysis of travel and stress caused by COVID-19 using TEXTOM, a big data analysis tool, and conducted semantic network analysis using UCINET6. We also conducted a CONCOR analysis to classify keywords for clustering of words with similarities. However, since we have collected travel and stress-oriented data from the start to the present, we need to increase the number of analysis data and analyze more data in the future.