• Journal of Nutrition and Health
• /
• v.41 no.5
• /
• pp.414-427
• /
• 2008

The purpose of the study was to develop an eating habit checklist for screening elementary school children at high risk of energy overintake. Dietary habits, food intake, anthropometric data were collected from 142 children (80 boys and 62 girls) in the 4th to 6th grades of elementary schools. Energy intake, fat intake, and percentage of Estimated Energy Requirement (%EER) were used as indices to detect the risk of energy overintake of the children. Pearson correlation coefficients were calculated between dietary habit scores and energy overintake indices in order to select questions included in the checklist. TV watching during the meal, meal speed, meal amount, overintake frequency, eatingout frequency, snack frequency, frequency of eating Ramyun or fast foods showed significant correlations with energy overintake indices. Stepwise regression analysis was performed to give each item a different weight by prediction strength. To determine the cut-off point of the test score, sensitivity, specificity, and positive predictive values were calculated. The 7-item checklist with test results from 0 to 13 points was developed, and those with equal or higher than 5 points were diagnosed as a risk group of energy overintake. Among our subjects 13.4% was diagnosed as the risk group. Mean energy intake of the subjects in the risk group and the normal group were 2,650 kcal and 1,640 kcal, respectively. However, there were no significant differences of Index of Nutritional Quality (INQ) of the other nutrients except eating fiber between the risk group and the normal group. This checklist will provide a useful screening tool to identify children at high risk of energy overintake.

• Knowledge Management Research
• /
• v.25 no.2
• /
• pp.1-18
• /
• 2024

With the increasing demand for and utilization of large language models (LLMs), the risk of user sensitive information being inputted and leaked during the use of LLMs also escalates. Typically recognized as a tool for mitigating the hallucination issues of LLMs, knowledge graphs, constructed independently from LLMs, can store and manage sensitive user information separately, thereby minimizing the potential for data breaches. This study, therefore, presents a knowledge graph-based chatbot that transforms user-inputted natural language questions into queries appropriate for the knowledge graph using LLMs, subsequently executing these queries and extracting the results. Furthermore, to evaluate the functional validity of the developed knowledge graph-based chatbot, performance tests are conducted to assess the comprehension and adaptability to existing knowledge graphs, the capability to create new entity classes, and the accessibility of LLMs to the knowledge graph content.

• Environmental and Resource Economics Review
• /
• v.7 no.2
• /
• pp.171-200
• /
• 1998

천연가스의 소비실태를 조사하기 위하여 실시한 표본조사자료를 이용하여 가정용 천연가스의 소비가 실증적으로 필수재의 성격을 가지는지를 조사하였다. 이를 위해 가장 일반적인 개념인 천연가스 소비의 소득에 대한 탄력성을 추정하였고, 다른 에너지와의 상대적인 비교를 위하여 광열비 지출에서의 천연가스 소비비 중의 소득에 대한 민감도를 추정하여 광열비 일반에 대한 천연가스 소비의 상대적 필수재의 성격을 파악하였고, 가정용 천연가스의 용도인 취사용과 난방용의 상대적 필수재 특성을 살피기 위하여 가스의 위험인지도에 대한 천연가스의 용도별 반응을 추정하였다. 이로부터 얻은 결과는 다음과 같다. 우선, i) 가정용 천연가스의 소비는 일반적으로 용도에 관계없이 필수재의 성격을 지니고 있으며, ii) 다른 에너지원과 비교를 위한 광열비에서의 천연가스 소비비 중의 소득 변화에 대한 반응을 보았을 때, 모형에 따라 다르나, 취사용이나 난방용이나 광열비 일반에 비하여 필수재 성격을 가진 것으로 나타나고 있으며, iii) 취사용 천연가스의 소비가 상대적으로 난방용에 비하여 필수재의 성격을 가진 것으로 나타나고 있다.

• Proceedings of the KIEE Conference
• /
• 2005.07a
• /
• pp.849-851
• /
• 2005

전력산업 구조개편 이후 여러 가지 형태의 새로운 전력시장 참여자가 나타났다. 지역 에너지 사업(CES : Community Energy Service) 사업도 이러한 새로운 시장참여자의 하나로 소비자에게 전기뿐만 아니라 열과 냉, 난방 서비스를 동시에 제공하는 역할을 수행한다. 대부분 분산전원 기술을 이용하는 지역 에너지 사업은 기존 발전 사업자들이 경험하는 환경문제나 대규모 투자에 따른 위험을 피할 수 있는 새로운 전력사업 형태로 각광받고 있다. 또한 송전망의 부담을 주지 않고 소비지에 인접하여 전력을 공급할 수 있기 때문에 송전망 투자도 절감할 수 있는 장점을 보인다. 본 연구는 지역 에너지 사업의 편익을 알아보기 위하여 부하 모델링을 통하여 지역 에너지 사업의 운영 경제성을 평가하였다. 특히 연료인 천연가스 요금과 전기부하의 변화에 따른 경제성 변화를 평가하여 민감도를 분석하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.624-626
• /
• 2022

최근 산업 현장에도 작업자들의 안전사고를 방지하기 위하여 인공지능 기법을 활용한 안전관리 시스템들이 도입되었다. 그러나 기존의 인공지능 기법을 활용한 방식은 데이터가 중앙 서버에 집중된 형태로 데이터 이동시 작업자의 민감 정보에 대한 보호가 어려울 뿐 아니라 대량의 데이터 발생 시 전체 시스템에 장애가 발생할 수 있어 이는 빠른 대응 프로세스가 필요한 산업 현장에 큰 영향을 줄 수 있다. 본 논문에서는 연합학습 기법을 적용하여 중앙 서버의 스트레스를 낮추어 작업자의 위험 상황에 빠른 대응이 가능하고, 작업자의 헬스 케어 데이터 같은 작업자의 민감 정보도 보호할 수 있는 시스템 설계를 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.681-683
• /
• 2023

폐암은 전 세계적으로 사망률이 가장 높은 암 질환으로, 조기 발견 및 신속한 치료를 위해서는 흉부 X-ray 영상 내 악성 결절을 놓치지 않는 것이 중요하다. 그러나 흉부 X-ray 영상은 정밀도의 한계로 진단 결과에 대한 신뢰도가 낮아, 이를 보조하는 도구의 개발이 요구된다. 기존의 폐암 진단 보조 도구는 학습 기반의 기법으로, 진단 결과에 대한 설명성(explainability)이 없다는 위험성을 갖는다. 이에 본 논문에서는 통계 분석에 기반한 결절의 석회화 여부 진단 기법을 제안한다. 제안하는 기법은 결절과 해부학적 구조물의 밝기 차 분포로부터 석회화 여부를 판단하며, 그 결과 민감도 65.22%, 특이도 88.48%, 정확도 83.41%의 성능을 보였다.

• Fire Science and Engineering
• /
• v.19 no.1 s.57
• /
• pp.20-28
• /
• 2005

Major petrochemical companies in the USA and the EU map out the strategies step-by-step hazard evaluation for the efficient risk management. They adopted the risk screening methods, such as Dow fire & explosion index, as a preliminary phase to execute detailed evaluation such as QRA (Quantitative Risk Assessment). In this study, The FEDI (Fire & Explosion Damage Index), which a kind of risk screening method proposed by Khan and Abbasi, was applied to the BTX plant in Korea. We showed that the FEDI can be effectively used to classify the hazard potential by comparison of the result from the FEDI and the result from QRA. And we showed that the characteristics and the quantities of chemical are the factors which have a largest effect on fire and explosion by executing relative sensitivity analysis of the FEDI. In conclusion, if the FEDI was applied as a preliminary phase of HAZOP, more efficient hazard evaluation can be possible.

• Journal of Korean Society of Transportation
• /
• v.25 no.5
• /
• pp.161-172
• /
• 2007

Vehicles transporting hazardous materials can make huge damage to people, properties and environment by traffic accidents. Therefore, transporting hazardous materials is a big issue with the cutting edge technology of communications in these days. However, despite this situation, Korean government gives limited efforts for systematic management, research and investment about hazardous materials. Accordingly, this research suggests the key path finding algorithm about management of real-time schedule and routes for vehicles transporting hazardous materials. Besides, the case study is progressed in transportation networks of Seoul in order to evaluate the reality of algorithm. Specifically, time-space network transformation is performed for time window attributes. In addition, this study proposes the techniques searching for non-dominated paths considering schedule by the multiobjective shortest path algorithm based on dynamic programming in dynamic transportation networks including multiobjective attributes.

• Journal of the Korean Society of Radiology
• /
• v.16 no.3
• /
• pp.309-317
• /
• 2022

It is reported that social distancing caused by the corona pandemic has resulted in a decrease in physical activity, and a decrease in muscle mass due to this increases the risk of vascular disease. Therefore, in this study, the risk of carotid intima media thickness was evaluated by measuring the thickness of CIMT, and the differences in variables according to gender were analyzed. From January 2022 to May 2022, a total of 220 people (121 males, 99 females) who performed blood tests, carotid ultrasound, and body composition analysis among those who visited the health check-up center of Busan H Hospital were included in this study. Significant differences in risk factors according to gender were analyzed using the chi-square test. The cut-off values of variables that can predict the risk of carotid intima media thickness were calculated, and the accuracy was evaluated by calculating the area under the curve, sensitivity, and specificity. As a result of the difference analysis, the higher the total cholesterol in men and the appendicular skeletal muscle index in women, the higher the risk of carotid intima media thickness. The cut-off value for predicting the risk of carotid intima media thickness was calculated to be greater than 199 mg/dL of total cholesterol in the male group and less than 5.9 kg/m² of the appendicular skeletal muscle index in the female group. As a result of this study, the higher the total cholesterol in men and the lower the appendicular skeletal muscle index in women, the higher the risk of increased CIMT. Therefore, in women, the appendicular skeletal muscle index is expected to serve as an indicator to predict and prevent vascular changes at an early stage.

• Journal of Nutrition and Health
• /
• v.43 no.2
• /
• pp.189-196
• /
• 2010

Malnutrition has been associated with higher hospital costs, mortality, rates of complications and longer length of hospital stay. Several nutritional screening tools have been developed to identify patients with malnutrition risk. However, many of those require much time and labor to administer and may not be applicable to a Korean population. Therefore, the aim of this study was to develop nutritional screening tool for Korean inpatients. Then we compare nutritional screening tools that developed and previously described. Seven hundred sixty-four patients at hospital admission were screened nutritional status and classified as well nourished, malnutrition stage 1 or stage 2 by the KNNRS (Kyunghee Neo Nutrition Risk Screening), PG-SGA (Patient-Generated Subjective Global Assessment) and NRS-2002 (Nutritional Risk Screening-2002). The KNNRS, PG-SGA and NRS-2002 respectively classified 28.7%, 51.3%, 48.5% of patients as malnourished status. Compared to the PG-SGA, the KNNRS had sensitivity 60.7% (95% CI 54.2-67.0) and specificity 81.2% (95% CI 75.3-85.2). Agreement was fair between KNNRS and PG-SGA (k = 0.34). Compared to the NRS-2002, the KNNRS had sensitivity 57.8% (95% CI 53.4-60.9) and specificity 64.4% (95% CI 60.2-69.8). Agreement was poor between KNNRS and NRS-2002 (k = 0.18). These result should include that the KNNRS and PGSGA have clinical relevance and fair concordance. However the rate of malnourished patients by KNNRS were less than by PG-SGA. For more effectivity of nutritional screening and management, the criteria of KNNRS would be better revised.