• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.71-73
• /
• 2022

최근 인프라를 관리할 필요가 없고 폭발적으로 늘어나는 요청을 유연하게 대처할 수 있는 장점 때문에 서버리스 컴퓨팅 사용이 늘어나고 있다. 하지만 서버리스 컴퓨팅은 사용자 코드의 실행 환경을 준비하기 위한 콜드 스타트 과정이 필요하고, 서비스가 복잡해짐에 따라 전체 실행 시간 중 콜드 스타트로 인한 지연시간이 늘어나는 문제가 발생한다. 본 논문에서는 서버리스 컴퓨팅 기반의 워크플로우에 대해 콜드 스타트로 인한 지연 시간을 완화하는 아키텍처 및 기법을 제안한다.

• Journal of Hydrogen and New Energy
• /
• v.33 no.1
• /
• pp.77-84
• /
• 2022

The experimental work has been carried out for the study of pyrolysis of oil samples used in industrial and utility boilers in Korea. For five oil samples, the characteristics of pyrolysis have been investigated with a thermogravimetric analyzer (TGA), and their kinetic parameters were obtained and compared each other. The rate order of pyrolysis rate for five oils were as follows: by-product fuel oil, pyrolysis oil, diesel, a heavy oil and refined oil. The pyrolysis of refined oil has been successfully described by the three step, first order reaction model while the single step reaction model has been used for other oils. For the reaction temperature over 550 K, the reactivity of refined oil was very poor compared with other oils.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.514-518
• /
• 2019

기존의 교차 도메인 추천은 일반적으로 서로 다른 도메인 데이터의 지식 결합이나 지식 공유를 바탕으로 진행된다. 이러한 방식들은 최소 한 개 이상의 도메인 데이터가 필요해서 모든 도메인의 피드백 데이터가 없는 실제 서비스 초기 상황에는 적합하지 않을 수 있다. 따라서 본 논문에서는 서비스 초반 모든 도메인의 피드백 데이터가 없고 콘텐츠 데이터만 존재하는 상황에서 교차 도메인 추천 시스템을 효과적으로 시작하기 위해 텍스트 임베딩, 클러스터링, 프로파일링 및 콘텐츠 기반 필터링을 활용한 추천 시스템 구성을 제안하고자 한다. 평가를 위해 여행지, 지역 축제, 공연을 포함하는 문화 관광 데이터와, 이에 대한 사용자 프로파일링 결과를 바탕으로 추천을 진행하였다. 그 결과, 콘텐츠 임베딩에 대한 유사도를 시각화하여 교차 도메인 아이템 간 유사성을 확인할 수 있었고, 사용자별 추천 결과를 통해 제안한 교차 도메인 추천 시스템이 유의미하게 동작함을 보였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.19-22
• /
• 2022

서비스 관점에서 구축되는 추천 시스템의 성능은 얼마나 효율적인 추천 모델을 적용하여 심층적으로 설계되었는가에 좌우된다고도 볼 수 있다. 특히, 추천 시스템의 초개인화는 세계적인 추세로 1~2년 전부터 구글, 아마존, 알리바바 등의 데이터 플랫폼 강자들이 경쟁적으로 딥 러닝 기반의 알고리즘을 개발, 자신들의 추천 서비스에 적용하고 있다. 본 연구는 갈수록 고도화되는 추천 시스템으로 인해 발생하는 여러 문제들 중 사용자 또는 서비스 정보가 부족하여 계속적으로 발생하고 있는 Cold-start 문제와 추천할 서비스와 사용자는 지속적으로 늘어나지만 실제로 사용자가 소비하게 되는 서비스의 비율은 현저하게 감소하는 데이터 희소성 문제 (Sparsity Problem)에 대한 솔루션을 모색하는 알고리즘 관점에서 연구하고자 한다. 본 논문은 첫 단계로, 적용하는 메타데이터에 따라 추천 결과의 정확성이 얼마나 차이가 나는지를 보이고 딥러닝 비지도학습 방식을 메타데이터 선정 및 추출에 적용하여 실시간으로 변화하는 소비자의 실제 생활 패턴 및 니즈를 예측해야 하는 필요성에 대해서 기술하고자 한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.59-60
• /
• 2022

갈수록 개인화되어 가는 추천시스템은 다양한 모델에 의해 그 성능이 향상되고 있으며 최근 추세는 다른 분야와 마찬가지로 딥러닝 기반 모델을 적용하여 추천 품질을 향상하고 있다. 그러나 대다수의 추천시스템은 하나의 도메인에서 개별적으로 사용될 뿐, 유사도메인이나 상이한 도메인이나 모두 다른 도메인에서의 사용자 성향이나 아이템 유사성을 거의 또는 전혀 고려하지 않고 있다. 이는 추천결과의 sparsity와 cold-start 문제를 더 악화시키는 원인이 된다. 본 논문은 다양한 딥러닝 모델 적용 추천 모델 중 오토인코더 모델을 지역특화 협업에 적용한 모델을 간략하게 소개하고 해당 모델을 상이한 도메인 간의 적용하기 위한 첫 단계로 손실함수 부분에 대해 개념적으로 설명하고자 한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.5
• /
• pp.1260-1272
• /
• 2024

This paper tackles the prevalent challenges faced by existing tourism route recommendation methods, including data sparsity, cold start, and low accuracy. To address these issues, a novel intelligent tourism route recommendation method based on collaborative filtering is introduced. The proposed method incorporates a series of key steps. Firstly, it calculates the interest level of users by analyzing the item attribute rating values. By leveraging this information, the method can effectively capture the preferences and interests of users. Additionally, a user attribute rating matrix is constructed by extracting implicit user behavior preferences, providing a comprehensive understanding of user preferences. Recognizing that user interests can evolve over time, a weight function is introduced to account for the possibility of interest shifting during product use. This weight function enhances the accuracy of recommendations by adapting to the changing preferences of users, improving the overall quality of the suggested tourism routes. The results demonstrate the significant advantages of the approach. Specifically, the proposed method successfully alleviates the problem of data sparsity, enhances neighbor selection, and generates tourism route recommendations that exhibit higher accuracy compared to existing methods.

• Journal of Korean Society for Atmospheric Environment
• /
• v.24 no.1
• /
• pp.72-82
• /
• 2008

Benzene is a very harmful and toxic compound known as human carcinogen by all routes of exposure. Owing to the risky feature of benzene, several countries such as Japan, UK and EU have established the ambient air quality standard and protect from that risk of it. Korea also has designated it as one of the criteria air pollutants and established the concentration limit ($5\;{\mu}g/m^3$) in the air and is going to apply the standard from 2010. Benzene is emitted from various sources such as combustion plants, production processes, waste treatment facilities and also automobiles. Mobile source is known as one of the major emission sources of benzene. In this study, we estimated the domestic emissions of benzene from mobile source and compared the results with those of advanced countries. Mobile source was divided into 2 categories, Le., on-road source and non-road source. The total emissions of benzene from mobile source were estimated as 3,106 tons/yr and 1,612 tons/yr was emitted from on-road source and 1,494 tons/yr was from non-road source. Emission ratio of benzene from on-road source showed that 80.0% was from passenger cars, 10.1% was from taxis, 7.2% was from light-duty vehicles, 2.5% was from heavy-duty vehicles and 0.2% was from buses. In the case of non-road source, the distribution showed that 66.3% was from construction machineries, 14.5% was from locomotives, 11.7% was from ships, 7.1% was from agriculture equipments and 0.5% was from aircrafts. The cold-start emissions were estimated as 942 tons/yr and this value was almost 1.5 times greater than that for hot engine emissions (608 tons/yr). In addition, the fuel-based distribution was 65.9%, 31.1% and 2.8% from gasoline, LPG and diesel vehicles, respectively. The emission ratio from mobile source occupied 65% and 30% of total benzene emissions in USA and UK, respectively. In case of Korea, the emission ratio of benzene from mobile source occupied 29% (15% from on-road source, 14% from non-road source) which showed similar value with UK.

• Journal of IKEEE
• /
• v.18 no.3
• /
• pp.420-426
• /
• 2014

In this paper, we have implemented the low temperature fluidity test system with the composited fuel filter and heater, which has tested the low temperature fluidity of gasoline, an engine start time, the consumption of electricity and power to evaluate the system performance. The test condition have used the diesel fuel, the normal temperature, the diesel fuel supply pressure $3.4kgf/cm^2$ at $-20{\sim}-30^{\circ}C$, the fuel supply quantization 60 l/H, the setting current 30 A and the voltage $13V_{dc}$. In order to simulate the operation of diesel fuel filter, we will establish the composited fuel filter into test jig, and be filled with chamber tank and filter by a constant flow quantization and pressure. After these, it shall be cold for setting time. And then we have measured the consumption current and power of heater, an operating time and pressure of filter.

• 한국해양학회지
• /
• v.25 no.2
• /
• pp.84-95
• /
• 1990

Observations by CTD castings, moored current meters and satellite imageries reveal some physical characteristics of the area around the tidal mixing front found in the mid-Yellow Sea off Korea. Tidal mixing is the greatest at the promontory of Taean Peninsula with a front around it. The front appears in April with the start of solar heating, becomes most clear in August and disappears in November with the start of surface cooling. In the north of the front, tidal fluctuations of temperature and salinity induced by tidal currents manifest the existence of the front, Differently from the usual tidal mixing front, the front in Kyunggi Bay is formed by presence of the water discharged from the Han River which meets the offshore water at the front. Near the surface cold center, vertically well-mixed zone extends to about 50 Km offshore from the coast, Farther south, this structure is generally retained but with lesser degree of vertical mixing. Within the relatively well-fixed coastal zone, the fresh water discharged from the Kum River makes another salinity front of smaller extent. At some places around this salinity front, an Upwelling-like feature is remarked.

• Physical Therapy Korea
• /
• v.3 no.2
• /
• pp.49-54
• /
• 1996

The purposes of this study were to measure skin temperature and blood flow on the contralateral upper extremity when heat is applied to one upper extremity, were to compare the effect of contrast bath. The subjects were 38 healthy adults with no history of peripheral vascular disease. The subjects of contrast bath were 18 persons and the subjects of warm bath were 20 persons. The subjects of one group were seated with their right arm in water($42^{\circ}C$) up to the mid-forearm. The subjects of the other group were seated with their right arm up to the mid-forearm in water which was changed from warm to cold using the contrast bath technique. The continually changing temperatures and blood flow were measured by an independent observer at intervals of 10, 15, 20, 25 and 30 minutes respectively after the start of the procedure. The results were as follows. The temperature of the warm bath group rose 4.28% over the pre-experimental temperatures and the temperature of the contrast bath group rose 3.41%. There was no statistically significant difference between the two groups. The blood flow of the warm bath group rose 8.31% over the pre-experimental blood flow and the blood flow of the contrast bath group rose 17.24%. There was a statistically significant between the two groups 20 minutes after the start of the procedure. Thus the contrast bath is a more effective method than the warm bath to increase blood flow.