• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.164-170
• /
• 2023

With the recent advances in AI (artificial intelligence) and HPC (high-performance computing) technologies, deep learning is proliferated in various domains of the 4th industrial revolution. As the workload volume of deep learning increasingly grows, analyzing the memory reference characteristics becomes important. In this article, we analyze the memory reference traces of deep learning workloads in comparison with traditional workloads specially focusing on read and write operations. Based on our analysis, we observe some unique characteristics of deep learning memory references that are quite different from traditional workloads. First, when comparing instruction and data references, instruction reference accounts for a little portion in deep learning workloads. Second, when comparing read and write, write reference accounts for a majority of memory references, which is also different from traditional workloads. Third, although write references are dominant, it exhibits low reference skewness compared to traditional workloads. Specifically, the skew factor of write references is small compared to traditional workloads. We expect that the analysis performed in this article will be helpful in efficiently designing memory management systems for deep learning workloads.

• Communications of the Korean Mathematical Society
• /
• v.11 no.3
• /
• pp.841-853
• /
• 1996

This article finds smoothed perturbation analysis estimates for the derivatives of mean performance measures in a Markovian queue with batch arrivals. We show that those estimates can be observed from a single sample path.

• Journal of the Korean Magnetic Resonance Society
• /
• v.22 no.3
• /
• pp.46-53
• /
• 2018

NMR-based metabolomics needs various knowledge to elucidate metabolic perturbation such as NMR experiments, NMR spectrum processing, raw data processing, metabolite identification, statistical analysis, and metabolic pathway analysis regarding technical aspects. Among them, some concepts of raw data processing and multivariate analysis are not easy to understand but are important to correctly interpret metabolic profile. This article introduces NMR spectrum processing, raw data processing, and multivariate analysis.

• Journal of the Korean Society for information Management
• /
• v.26 no.3
• /
• pp.377-394
• /
• 2009

There are some limitations in the existing bibliometric methods to satisfy the various requests of the interest parties including researchers, managers, policy makers to identify 1) which research group or researcher is the key player, and the overall trends of the particular technological sub-fields, 2) which research groups, institutions or countries mainly use their research outputs, 3) what are the spin-offs from research outputs to some scientific and technological fields, 4) in which levels they are when comparing their quantitative and qualitative research outputs to those of other competitive institutions. It is essential to develop new informetric indicators and methodologies in order to satisfy stakeholder's various demands and to strengthen qualitative analysis in measuring research performance. This study suggested informetric indicators such as article quality index, citation impact index, international cooperation index, excellent article production index and methodologies including citation analysis, text mining.

• Journal of Families and Better Life
• /
• v.26 no.5
• /
• pp.177-194
• /
• 2008

The purpose of this article was to be offered data for the efficient management of healthy family support center as the deduction of the standard job of healthy family supporter. The process of this study take the major steps as subject selection based on career, investigation about range and contens of healthy family support work and deduction and examination job areas, duty and task. Consequently, the job areas of healthy family supporter are generalization, counseling, education, culture, administrative affairs, taking care of children and taking care of children for family living with a handicapped child. The standards of job analysis are frequency, importance and number of human power. The result and procedure of this article provides the main data and idea for the development of a tool of measurement, ajob analysis and information of the specialty and role of healthy family supporter.

• Computers and Concrete
• /
• v.21 no.1
• /
• pp.11-20
• /
• 2018

The possibilities of non-linear analysis of reinforced-concrete structures are under development. In particular, current research areas include structural analysis with the application of advanced computational and material models. The submitted article aims to evaluate the possibilities of the determination of material properties, involving the tensile strength of concrete, fracture energy and the modulus of elasticity. To evaluate the recommendations for concrete, volume computational models are employed on a comprehensive series of tests. The article particularly deals with the issue of the specific properties of fracture-plastic material models. This information is often unavailable. The determination of material properties is based on the recommendations of Model Code 1990, Model Code 2010 and specialized literature. For numerical modelling, the experiments with the so called "classic" concrete beams executed by Bresler and Scordelis were selected. It is also based on the series of experiments executed by Vecchio. The experiments involve a large number of reinforcement, cross-section and span variants, which subsequently enabled a wider verification and discussion of the usability of the non-linear analysis and constitutive concrete model selected.

• The Journal of Korean Medical History
• /
• v.29 no.1
• /
• pp.47-54
• /
• 2016

This article attempts to extract headwords for complication of "Donguibogam Dictionary" with Corpus-based Analysis. The computerized original text of Donguibogam is changed into a text file by a program 'EM Editor'. Chinese characters of high frequency of exposure among Chinese characters of Donguibogam are extracted by a Corpus-based analytical program 'AntConc'. Two-syllable, three-syllable, four-syllable, and five-syllable words including each Chinese characters of high frequency are extracted through n-cluster, one of functions of AntConc. Lastly, The output that is meaningful as a word is sorted. As a result, words that often appear in Donguibogam can be sorted in this article, and the names of books, medical herbs, disease symptoms, and prescriptions often appear especially. This way to extract headwords by this Corpus-based Analysis can suggest better headwords list for "Donguibogam Dictionary" in the future.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.8
• /
• pp.387-397
• /
• 2021

The object of this article is to assess the factors affecting the behavioral intention of young retail customers to use digital banking services in Vietnam. In this article, multivariate data analysis techniques including Cronbach's Alpha, Exploratory factor analysis (EFA), Confirmatory factor analysis (CFA), Structure equation model (SEM), and Bootstrapping are used to analyze the data collected from 525 young respondents under the age of 35 who are using or having opportunities to experience digital banking services. The people taking part in the survey were mainly University students with incomes of most of them under VND 5 million. The result from the analysis illustrates that (1) perceived ease of use positively affects intention to use, (2) social influence positively impacts intention to use, and (3) customer support has a positive impact on the intention of young people to use digital banking services. While technology self-efficacy, convenience, and perceived security were found to have an impact on intention to use services in former studies, the influences of these factors on intention to use digital banking services are found insignificant in this research. From these results, the author provides implications for commercial banks to increase the intention to use digital banking of young people in Vietnam.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.10
• /
• pp.176-182
• /
• 2018

In order to understand and track the trends of construction safety accident, this study shows the topic trends in the construction safety accident with LDA(Latent Dirichlet Allocation)-based topic modeling method for data analytics. Especially, it performs to figure out the main issue of construction safety accident with unstructured data analysis based on the topic modeling rather than a variety of structured data analysis for preventing to safety accident in construction industry. To apply this methodology, I randomly collected to 540 news article data about construction accident from January 2017 to February 2018. Based on the unstructured data with the LDA-based topic modeling, I found the 10 topics and identified key issues through 10 keyword in each 10 topics. I forecasted the topic issue related to construction safety accident based on analysis of time-series trends about the news data from January 2017 to February 2018. With this method, this research gives a hint about ways of using unstructured news article data to anticipate safety policy and research field and to respond to construction accident safety issues in the future.

• Journal of Arbitration Studies
• /
• v.22 no.2
• /
• pp.103-124
• /
• 2012

Because the Korea-U.S. Free Trade Agreement (Korea-U.S. FTA) and the North American Free Trade Agreement (NAFTA) have an overlapping contracting party, the United States, their provisions have much in common. The investment chapters of these agreements, especially, show many similarities, and thanks to these similarities, it is likely that the Korea-U.S. FTA arbitration tribunal for investor-state disputes regarding the environment will put great weight on the NAFTA tribunals' interpretations of those similar provisions. Since the NAFTA tribunals have already handled many environment-related arbitration cases, their interpretations will help heighten the predictability of environment-related Korea-U.S. FTA arbitration cases. This paper analyzes the environment-related NAFTA cases in which the tribunal has issued an award, which are the Metalclad case, S.D. Myers case, Waste Management case, Methanex case, Glamis Gold case, and Chemtura case. According to this analysis, the most controversial NAFTA provisions have been Article 1102 (national treatment), Article 1105 (minimum treatment standard, fair and equitable treatment), and Article 1110 (expropriation). The NAFTA tribunals applied the requirement of these articles in a strict manner, reducing the possibility of finding a violation. After the aforementioned analysis, this paper proceeds to compare the national treatment, minimum treatment standard (fair and equitable treatment), and expropriation provisions of the Korea-U.S. FTA and NAFTA and to predict the impact that the environment-related awards under NAFTA can have on environment-related Korea-U.S. FTA cases. It is expected that the NAFTA interpretations of the national treatment and minimum treatment provisions are likely be used as they are, but not the interpretations of expropriation, because of the differences in the expropriation provisions of the two agreements.