• Journal of The Korean Association of Information Education
• /
• v.25 no.5
• /
• pp.813-824
• /
• 2021

To improve educational performance by analyzing the learning trends of adult learners of Open High Schools, various algorithm models using artificial intelligence were designed and performance was evaluated by applying them to real data. We analyzed Log data of 115 adult learners in the cyber education system of Open High Schools. Most adult learners of Open High Schools learned more than recommended learning time, but at the end of the semester, the actual learning time was significantly reduced compared to the recommended learning time. In the second half of learning, the participation rate of VODs, formation assessments, and learning activities also decreased. Therefore, in order to improve educational performance, learning time should be supported to continue in the second half. In the latter half, we developed an artificial intelligence algorithm models using Tensorflow to predict learning time by data they started taking the course. As a result, when using CNN(Convolutional Neural Network) model to predict single or multiple outputs, the mean-absolute-error is lowest compared to other models.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.110-118
• /
• 2021

The Internet of Things (IoT) technology, a key growth engine of the 4th industrial revolution, has grown to a stage where it can autonomously communicate with each other and process data according to space and circumstances. Accordingly, the IT infrastructure becomes increasingly complex and the importance of the monitoring field for maintaining the system stably is increasing. Monitoring technology has been used in the past, but there is a need to find a flexible monitoring system that can respond to the rapidly changing ICT technology. This paper conducts research on designing and testing an open source-based Prometheus monitoring system. We builds a simple infrastructure based on IoT devices and collects data about devices through the Exporter. Prometheus collects data based on pull and then integrates into one dashboard using Grafana and visualizes data to monitor device information.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.118-118
• /
• 2022

지속적인 학문의 발전을 위해서는 선행연구에 대한 재현성이 무엇보다도 중요하다고 할 수 있다. 하지만 컴퓨터와 소프트웨어의 급속한 발달로 인한 컴퓨터 환경의 다양화, 분석 소프트웨어의 지속적 최신화로 인해서 최근 구축된 모델도 짧게는 몇 달, 길게는 1~2년후면 다양한 에러로 인하여 재현성이 불가능해지고 있다. 이러한 재현성의 극복을 위해서 온라인을 통한 데이터와 소스코드의 공유의 필요성이 제시되고 있으나, 실제로는 개인마다 컴퓨터 환경, 버전, 소프트웨어 설치에 필요한 라이브러리의 버전 또는 디렉토리 등이 달라 단순히 온라인을 통한 데이터와 소스코드의 공유만으로 재현성을 개선하기는 힘든 것이 현실이다. 따라서 이러한 컴퓨터 모델링 환경의 공유는 과거의 형태와 같이 데이터, 소스코드와 매뉴얼의 공유만으로 불가능하다고 할 수 있다. 따라서 본 연구에서는 수자원 모델링의 재현성 개선을 위해 1) 온라인 저장소, 2) 클라우드기반 JupyterHub 모델링 환경과 3) 모델 APIs 3개의 핵심 구성요소를 제시하고, 최근 미국에서 개발된SUMMA(Structure for Unifying Multiple Modeling Alternative) 수자원 모델에 적용하여 재현성 달성을 위한 3개의 핵심 구성요소의 필요성과 용이성을 검증하였다. 첫 번째, 데이터와 모델의 온라인 공유는 FAIR(Findable, Accessible, Interoperable, Reusable) 원칙으로 개발된 수자원분야의 대표적인 온라인 저장소인 HydroShare를 활용하여 모델입력자료를 메타데이터와 함께 공유하였다. 두 번째, HydroShare에서 Web App의 형태로 제공되는 클라우드기반 JupyterHub환경인 CUAHSI JupyterHub(CJH)와 일루노이대학에서 제공하는 CyberGIS-Jupyter for water JupyterHub(CJW)환경에 수자원모델링 환경을 컨테이너(Docker) 환경을 통해 구축·공유하였다. 마지막으로, 클라우드에서 수자원모델의 효율적 이용을 위해 Python기반의SUMMA모델 API인 pySUMMA를 개발·공유하였다. 이와같이 구축된 3개의 핵심 구성요소를 이용하여 2015년 Water Resources Research에 게재된 SUMMA 논문의 9개 Test Cases 중에서 5개를 누구나 쉽게 재현할 수 있음을 증명하였다. 재현성의 중요성에 대한 인식의 증가로 Open과 Transparent Hydrology에 대한 요구가 증대되고 있으며, 이를 위해서 클라우드 기반의 모델링 환경구축 및 제공이 확대되고 있다. 본 연구에서 제시한 HydroShare와 같은 온라인 저장소, CJH와 CJW와 같은 클라우드기반 모델링환경, 모델의 효율적 이용을 위한 모델 APIs는 급속도로 발달하고 있는 컴퓨터 및 소프트웨어 환경에서 핵심구성요소이며, 연구의 재현성 개선을 통해 수자원공학 발전에 기여할 것으로 기대된다.

• Reproductive and Developmental Biology
• /
• v.41 no.2
• /
• pp.25-31
• /
• 2017

Effective estrus detection and artificial insemination (AI) are necessary for profitable management of dairy herd. In current study, 45 crossbred lactating cows have been selected with the complaint of unobserved oestrus for more than sixty days postpartum. All cows had functional corpus luteum as examined by transrectal ultrasonography. Cows were treated with $PGF_2{\alpha}$ analogue and AI was performed with observed oestrus and then single dose of GnRH was administered. Similar synchronization protocol has been repeated after 14 days in cows that did not repose to first treatment. Remaining cows received additional $PGF_2{\alpha}$ after 14 days of second treatment and timed AI was performed following GnRH administration. Among 45 cows, 28.89% showed estrus after first treatment and 78.79% responded to second hormonal intervention. A higher conception rate (88.89% vs 26.66 and 72.72%) was observed in cows after triple administration of $PGF_2{\alpha}$ and timed AI. We noticed a significant differences in body condition score (BCS, 1~5 scale), postpartum period, and daily milk production between cows that either responded of non-responded following first and second hormonal treatment. In addition, there was a significant positive correlation between daily milk production and BCS, age and postpartum days, milk production and estrus/BCS, and milk production/BCS/estrus and conception rate. Depending upon the findings we conclude that hormonal intervention with $PGF_2{\alpha}$ and GnRH enhances postpartum ovarian cyclicity and help decreasing the days open of dairy herd. Therefore, this finding might provide an excellent guideline for target breeding system for profitable dairy herd management.

• Economic and Environmental Geology
• /
• v.55 no.3
• /
• pp.309-316
• /
• 2022

In order to minimize the human and time consumption required for rock classification, research on rock classification using artificial intelligence (AI) has recently developed. In this study, basic volcanic rocks were subdivided by using polarizing microscope thin section images. A convolutional neural network (CNN) model based on Tensorflow and Keras libraries was self-producted for rock classification. A total of 720 images of olivine basalt, basaltic andesite, olivine tholeiite, trachytic olivine basalt reference specimens were mounted with open nicol, cross nicol, and adding gypsum plates, and trained at the training : test = 7 : 3 ratio. As a result of machine learning, the classification accuracy was over 80-90%. When we confirmed the classification accuracy of each AI model, it is expected that the rock classification method of this model will not be much different from the rock classification process of a geologist. Furthermore, if not only this model but also models that subdivide more diverse rock types are produced and integrated, the AI model that satisfies both the speed of data classification and the accessibility of non-experts can be developed, thereby providing a new framework for basic petrology research.

• Journal of Korean Library and Information Science Society
• /
• v.54 no.1
• /
• pp.45-68
• /
• 2023

This study aims to investigate the awareness and experiences of domestic researchers regarding data publishing, which has been recognized as a new channel of data sharing as scholarly communication evolves in the open science environment. A survey is conducted among researchers from five government-funded research institutes in the field of science and technology and members of the GeoAI Data Society to confirm the awareness of data publishing. As a result of the study, domestic researchers recognized providing explanations for data, stable access to data, citation, and quality assurance through peer review as the advantages of data journals. On the contrary, a low level of recognition for data paper as one of the research outputs was presented. With regard to the properties of data publication, the respondents answered that the data description, metadata description, and permanent identifiers are highly related, however, their recognition of the relation between the properties of data publication and the data submission to a repository and data peer review was relatively low. Finally, to expand the data publication, the need for the development of an editorial system that supports data paper peer review and cross-linking to a data repository as well as the development of a repository that supports data citation was identified. This study on the domestic researchers' experience and awareness of data publishing can provide insights for the implementation of data publishing services and infrastructure in the future.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.219-240
• /
• 2023

Conversational AI which allows users to interact with satisfaction is a long-standing research topic. To develop conversational AI, it is necessary to build training data that reflects real conversations between people, but current Korean datasets are not in question-answer format or use honorifics, making it difficult for users to feel closeness. In this paper, we propose a conversation dataset (KOMUChat) consisting of 30,767 question-answer sentence pairs collected from online communities. The question-answer pairs were collected from post titles and first comments of love and relationship counsel boards used by men and women. In addition, we removed abuse records through automatic and manual cleansing to build high quality dataset. To verify the validity of KOMUChat, we compared and analyzed the result of generative language model learning KOMUChat and benchmark dataset. The results showed that our dataset outperformed the benchmark dataset in terms of answer appropriateness, user satisfaction, and fulfillment of conversational AI goals. The dataset is the largest open-source single turn text data presented so far and it has the significance of building a more friendly Korean dataset by reflecting the text styles of the online community.

• Journal of Families and Better Life
• /
• v.23 no.6 s.78
• /
• pp.179-190
• /
• 2005

This study discusses the conceptual meanings of 'healthy families' by examining four popular misunderstandings regarding the concept. These misunderstandings are based partly on the lack of consensus on the use of the concepts of 'health' and 'families' and partly on the intentional misreading of the 'Healthy Fanulies Act' To correspond to the Concerns related to the Act, we need to clarify various meanings of the concept of family and to confirm the theoretical pounds of 'healthy families' based on the multidisciplinary consensus. To build consensus, it might be necessary to review some of the articles of the Act that have been misinterpreted.

• Journal of Families and Better Life
• /
• v.24 no.3 s.81
• /
• pp.59-72
• /
• 2006

This article explores bilateralization phenomena in Korean families through intimacy and interaction with matrilineal kins. In-depth interviews were conducted with married men and women aged from their mid-30s to mid-40s. The findings indicate that the intimacy range of participants was restricted to blood-based kins. The power hypothesis was supported for male participants. Some wanted to actively contact their spouse's family members, while others did not want to be connected with the spouse's families. Interaction with wives' families was based on the exchange perspective. In contrast, a complicated mechanism existed in the intimacy and interaction of female participants. Most of them were connected with in-laws in a passive way but actively interacted with their family-of-origin. However, some female respondents did not have good relationships with their mothers. Undifferentiation between mothers and married daughters negatively influenced their relationships. Therefore, it could be concluded that bilateralization phenomena in Korean families is another kind of shackle of patrilineal norms.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.50 no.11
• /
• pp.522-531
• /
• 2001

Recently conventional artificial intelligence(AI) approaches have been employed to build action selectors for the autonomous mobile robot(AMR). However, in these approaches, the decision making process to choose an action from multiple competence modules is still an open question. Many researches have been focused on the reactive planning systems such as the biological immune system. In this paper, we attempt to construct an action selector for an AMR based on the artificial immune network and internet. The information from vision sensors is used for antibody. We propose a learning method for artificial immune network using evolutionary algorithm to produce antibody automatically. The internet environment for an AMR action selector shows the usefulness of the proposed learning artificial immune network application.