• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1773-1793
• /
• 2023

Automated machine learning, often referred to as "AutoML," is the process of automating the time-consuming and iterative procedures that are associated with the building of machine learning models. There have been significant contributions in this area across a number of different stages of accomplishing a data-mining task, including model selection, hyper-parameter optimization, and preprocessing method selection. Among them, preprocessing method selection is a relatively new and fast growing research area. The current work is focused on the recommendation of preprocessing methods, i.e., feature subset selection (FSS) algorithms. One limitation in the existing studies regarding FSS algorithm recommendation is the use of a single learner for meta-modeling, which restricts its capabilities in the metamodeling. Moreover, the meta-modeling in the existing studies is typically based on a single group of data characterization measures (DCMs). Nonetheless, there are a number of complementary DCM groups, and their combination will allow them to leverage their diversity, resulting in improved meta-modeling. This study aims to address these limitations by proposing an architecture for preprocess method selection that uses ensemble learning for meta-modeling, namely AutoFE-Sel. To evaluate the proposed method, we performed an extensive experimental evaluation involving 8 FSS algorithms, 3 groups of DCMs, and 125 datasets. Results show that the proposed method achieves better performance compared to three baseline methods. The proposed architecture can also be easily extended to other preprocessing method selections, e.g., noise-filter selection and imbalance handling method selection.

• Journal of muscle and joint health
• /
• v.23 no.2
• /
• pp.105-113
• /
• 2016

Purpose: The purpose of this study was to investigate the effect of Situation-Background-Assessment-Recommendation (SBAR) reports on communication clarity and nurse-physician collaborative relationships. Methods: SBAR forms in one group were measured at baseline, at 2 weeks, and at 4 weeks postintervention. The subjects were 30 nurses who worked in a general ward of a general hospital in B city, Korea. A 45-minute SBAR education was provided to all subjects. Data was collected between September to October 2015. The collected data was analyzed using a repeated measure ANOVA. Results: After SBAR reports, levels of nurse's communication clarity and nurse-physician collaborative relationships were significantly higher than levels before SBAR reports. Conclusion: SBAR reports improved the communication clarity and nurse-physician collaborative relationships. It is recommended to use a lot of SBAR in clinical practice.

• Asian-Australasian Journal of Animal Sciences
• /
• v.6 no.4
• /
• pp.509-514
• /
• 1993

The experiment was conducted to determine the effect of different fiber levels on milk production of crossbred Holstein milking cows fed urea-treated rice straw (UTS) as a roughage. Eight cows were allotted into 2 squares of 4 cows each with 4 treatments by a balanced design. The treatments were 17%, 22%, and 24% crude fiber (CF) diets and Thai feeding system (free choice of roughage and 1 kg of concentrates/2 kg of milk) as a control. Body weight change was not significantly different among the treatments during the experiment. Milk production (4% FCM) and milk protein content wee not different among the treatments, but milk fat content was low in the 17% CF group and high in the control group. Cows fed the 17% CF diet consumed less UTS and more concentrates than the others, and consequently total DM intake was not different among the treatments. The feed conversion ratio was significantly higher in the control. Feed cost per kg milk was lowest in the control and highest in the 17% CF diet. The fiber content of the diet would be more than 17%, preferably 22-24% for normally producing Thai crossbred Holstein cows when the UTS was fed as a main roughage source.

• Women's Health Nursing
• /
• v.17 no.3
• /
• pp.205-214
• /
• 2011

Purpose: The purpose of this study was to examine womens' perceptions regarding influenza vaccination during pregnancy among Korean childbearing women. Methods: Data was collected using focus group interviews from June to September, 2010. Forty Korean women participated in 13 focus groups. After obtaining permission from participants, each session of the focus group was audio-taped and transcribed verbatim. The responses were analyzed utilizing qualitative content analysis. Results: Forty women ranging from 26 to 43 participated in the study. The major themes were safety concerns; effects of fetal protection and infection prevention; lack of perceived needs; and encouragement as well as concerns from others. Participants raised questions on whether the vaccine was safe and effective, and concerns about the potential harmful effect of influenza vaccine to their bodies and the fetus. The major reason for influenza vaccination during pregnancy was to protect self and fetus. Also, clinician's recommendation was the facilitating factors for influenza vaccination during pregnancy. Conclusion: The findings suggest that concerns associated with influenza vaccination should be considered when educating childbearing women about the influenza vaccine during pregnancy.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.175-177
• /
• 2006

According to the recommendation of International Maritime Organization, the Ministry Of Maritime Affairs and Fisheries (MOMAF) of Korea provides the real time Differential Global Positioning System service using maritime radio beacon from 1999. Due to the benefit of DGPS service, the need of this system is increased from various user groups for acquiring the better accuracy and integrity. Therefore, MOMAF has extended their service to inland by installing the additional 6 DGPS stations. This nationwide DGPS service will be fully deployed at 2007. In addition to the extension of service area, MOMAF has a plan to upgrade their nationwide DGPS to High Accuracy Nationwide DGPS (HANDGPS). The planned HANDGPS service of Korea will be a kind of long range RTK or Wide Area RTK techniques to provide under 1m accuracy and start their service from 2009 using the various broadcasting and communication media like as radio beacon, Wibro, Digital Multimedia Broadcasting, High Speed Packet Data Access. The introduction of nationwide DGPS system of Korea and its evolution plan will be addressed in this paper. The research activities related with HANDGPS in Korea is also presented.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.396-404
• /
• 2016

In this paper, the customized sport convergence contents curation system is proposed for activation of sport for all. The proposed system collects and analyzes profile of social sports group(club, society, etc.) for recommend optimized sport convergence contents to user. And the survey is progressed for research problem of existed service and usefulness of proposed service. The proposed system and service are expected to design innovative new service model for activate of sport for all.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1994.10a
• /
• pp.183-190
• /
• 1994

The paper provides a review of the analysis methods currently being used to perform seismic analysis of free standing high density spent fuel storage racks for PWR. On the basis of the analysis techniques obtained by KAERI from the design experience of Yonggwang unit 3&4 and Ulchin unit 3&4, the analysis procedure and modeling methods are discussed. The analysis of free standing fuel racks requires consideration of complex phenomena such as hydrodynamic coupling, impact through gap between fuel assembly and poison box and racks, frictional effect, rigid body sliding and tipping and etc. The present modeling of these factors is reviewed in comparison with the recommendation of regulatory group. Further improvement of analysis method and the current issues for the development are discussed.

• Journal of applied mathematics & informatics
• /
• v.28 no.1_2
• /
• pp.439-450
• /
• 2010

In this paper, we propose a new idea to evaluate the prediction accuracy of user's preference generated by memory-based collaborative filtering algorithm before prediction process in the recommender system. Our analysis results show the possibility of a pre-evaluation before the prediction process of users' preference of item's transaction on the web. Classification functions proposed in this study generate a user's rating pattern under certain conditions. In this research, we test whether classification functions select users who have lower prediction or higher prediction performance under collaborative filtering recommendation approach. The statistical test results will be based on the differences of the prediction accuracy of each user group which are classified by classification functions using the generative probability of specific rating. The characteristics of rating patterns of classified users will also be presented.

• Pediatric Infection and Vaccine
• /
• v.15 no.1
• /
• pp.1-4
• /
• 2008

Immunizations are among the most cost-effective and widely used public health interventions. This is a report a revision of recommendation of immunization for children by Korean Pediatric Society. Immunization. Vaccines were divided into 4 groups. 1) Vaccines that are recommended to all infants and children (BCG, hepatitis B vaccine, DTaP, Td, Polio vaccine, Japanese encephalitis vaccine, MMR, varicella vaccine, influenza vaccine [6-23 months of age], H. influenzae type b vaccine), 2) those that can be administered to all infants and children, but decision of administration is made by parents (pneumococcal conjugate vaccine, hepatitis A vaccine, influenza vaccine [healthy children ${\geq}24$ months of age], rotavirus vaccine, human papilloma virus vaccine), 3) those that should be given to high risk group (pneumococcal polysaccharide vaccine [high risk patients ${\geq}24$ months of age], influenza vaccine [high risk patients ${\geq}24$ months of age], typhoid vaccine), and 4) those administered for control of outbreaks or prevention of emerging infectious diseases. Immunization schedule recommended by Korean Pediatric Society in 2008 is presented.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10b
• /
• pp.169-173
• /
• 2006

유비쿼터스 상거래에서 사용자가 정보를 효율적으로 이용할 수 있도록 제어하고 필터링하는 일을 도와주는 개인화된 추천 시스템이 등장하였다. 더 나아가서는 사용자가 원하는 아이템을 예측하고 추천해주며, 이를 위해 협력적 필터링 기술을 적용하고 있다. 이는 사용자의 성향에 맞는 아이템을 예측하고 추천하기 위하여 비슷한 선호도를 가지는 사용자들 간의 유사도 가중치를 계산한다. 본 논문에서는 속성정보에 대한 사용자의 선호도를 고려하지 않은 문제점을 개선하기 위해서 속성정보를 이용한 연관 사용자의 선호도를 협력적 필터링 기술에 반영함으로써 추천의 정확도를 높이고자 한다. 그리고 협력적 필터링의 {연관 사용자-아이템} 행렬에서 사용자들 간의 연관 관계를 유지하면서 차원 수를 감소시키기 위해 ARHP 알고리즘을 이용하여 연관 사용자 군집을 한다. 제안된 방법의 성능 평가를 하기 위해 사용자가 아이템에 대해서 평가한 MovieLens 데이터 집합을 대상으로 평가되었으며, 기존의 Nearest Neighbor Model과 K-Means 군집보다 그 성능이 우수함을 보인다.