• Title/Summary/Keyword: 중립 모델

Search Result 180, Processing Time 0.028 seconds

Generative AI based Emotion Analysis of Consumer Reviews Using the Emotion Wheel (생성 AI 기반 감정 수레바퀴 모델을 활용한 사용자 리뷰 감정 분석)

  • Yu Rim Park;Hyon Hee Kim
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2023.11a
    • /
    • pp.1204-1205
    • /
    • 2023
  • 본 논문은 소비자의 리뷰 데이터를 기반으로 한 새로운 감성 분석 방법을 제안한다. 긍정, 부정, 중립으로 분류하는 전통적 감성 분석방법은 텍스트에 나타난 감정의 섬세한 차이를 파악하기 어렵다. 이에 본 연구에서는 GPT 모델을 사용하여 텍스트에서 사용자의 감정을 8 가지의 카테고리로 세분화한다. 부정적 정서를 가진 리뷰에서 분노, 혐오, 실망과 같은 구체적인 감정들을 직관적으로 파악할 수 있었고, 감정의 강도까지 파악할 수 있었다. 제안된 방법을 통해 기업은 고객의 요구 사항을 정확하게 인지할 수 있으며, 고객 맞춤형 서비스 개선에 기여할 수 있다는 점이 기대된다.

LSTM-based server management model for carbon-neutral data center operation (탄소중립적 데이터 센터 운영을 위한 LSTM기반 서버 관리 모델)

  • Ma, Sang-Gyun;Park, Jaehyun;Seo, Yeong-Seok
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2022.05a
    • /
    • pp.487-490
    • /
    • 2022
  • 최근 데이터 활용이 중요해지고 있는 시대인 만큼 데이터센터의 중요도도 높아지고 있다. 하지만 데이터센터는 24시간 가동되는 막대한 전력을 소모하는 시설이기 때문에 환경적, 경제적 측면에서 문제가 되고 있다. 최근 딥러닝 기법들을 사용하여 데이터센터나 서버에서 사용되는 전력을 줄이거나, 트래픽을 예측하는 연구들이 다양한 관점에서 이루어지고 있다. 그러나 서버에서 처리되는 트래픽 데이터양은 변칙적이며 이는 서버를 관리하기 어렵게 만든다. 또한, 가변적으로 서버를 관리하는 기법에 대한 연구들이 여전히 많이 요구되어지고 있다. 따라서 본 논문에서는 이러한 문제점을 해결하기 위해 시계열 데이터 예측에 강세를 보이는 장단기 기억 신경망(Long-Term Short Memory, LSTM)을 기반으로 한 가변적인 서버 관리 기법을 제안한다. 제안된 모델을 통해 현업환경에서 이전보다 안정적이고 효율적으로 서버를 관리할 수 있게 되며, 서버에서 사용되는 전력을 보다 효과적으로 줄일 수 있게 된다. 제안된 모델의 검증을 위해 위키 피디아(WikiPedia) 서버의 트래픽 데이터양을 수집한 뒤 실험을 수행하였다. 실험 결과 본 논문에서 제안된 모델이 유의미한 성능을 보이며, 서버 관리를 안정적이고 효율적으로 수행할 수 있음을 보여주었다.

Effects of the ANP Models on the Comparison Indicators of Electric Power Systems (ANP 모델이 전력 시스템의 비교 지표에 미치는 영향)

  • Kim Seong-Ho;Kim Kil-Yoo;Kim Tae-Woon
    • Proceedings of the Korea Society for Energy Engineering kosee Conference
    • /
    • 2006.05a
    • /
    • pp.371-376
    • /
    • 2006
  • 서로 갈등적인 관계에 있는 다중 기준 하에서 다양한 국가 전력 시스템을 정량적으로 비교하는 데에는 전력 시스템의 비교 지표가 필요하다. 이러한 비교 지표를 산출하기 위하여 해석적 망형 과정(Analytic Network Process; ANP) 모델 가운데 상호 의존도 수중이 낮은 되먹임 모델 및 상호 의존도가 없는 독립성 모델이 개발되었다. 이러한 ANP 모델은 구성요소로 교점들(nodes)과 상호작용 관계를 표현하는 가지들(arcs)을 포함하고 있다. 의사결정 목표 교점에는 세 가지 유형의 리스크 성향이 포함되었다. 이러한 리스크 성향은 원자력 발전소 같은 위험 설비에 대한 전문가(그룹)의 리스크 성향이며, 더 구체적으로 말하면, 리스크 감수 성향, 리스크 혐오 성향, 리스크 중립 성향 등이다. 여기서 수행된 연구의 주요 목적은 ANP 모델을 구성하는 교점들 가운데 하나인 평가 기준 교점에서의 변화가 전력 시스템의 비교 지표에 미치는 영향을 해석하려는 것이다. 이러한 모델 변이가 비교 지표에 미치는 영향을 알아보기 위한 사례 연구에서 각 발전원의 특성을 비교할 평가 기준은 기준 사례와 비교 사례 각각에 대하여 상이하게 선정되었다: 기준 사례의 경우에는 보건성을 대표하는 생명 단축 [yr/TWh], 환경성을 대표하는 지구 온난화 [$g\;CO_{2}-eq./kWh$], 사회성을 대표하는 지속가능 정도[-], 경제성을 대표하는 발전 단가 [\/kWh] 등이 선정되었다; 반면에, 비교 사례의 경우에서는 보건성을 대표하는 사고 사망 [death/GWh]만이 다르고 나머지는 동일하게 선정되었다. 이러한 보건성을 대표하는 생명 단축 또는 사고 사망의 선정은 다음과 같은 비교 지표에 영향을 미친다는 것이 발견되었다: (1) 되먹임 모델에서는 성향 가중치 및 기준 등급에 영향을 준다. (2) 되먹임 모델과 독립성 모델에서는 시스템 등급에 영향을 준다. 향후에는 더욱 더 다양한 상호의존 모델들이 정량화될 필요성이 있다고 본다.

  • PDF

The study related to the meta data for the attribute mapping from IFC to CityGML (IFC에서 CityGML로 속성 맵핑을 위한 메타 데이터에 관한 연구)

  • Kang, Tae Wook;Choi, Hyun Sang;Hwang, Jung Rae;Hong, Chang Hee
    • Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
    • /
    • v.30 no.6_1
    • /
    • pp.559-565
    • /
    • 2012
  • The purpose of the present study is to suggest the meta data of the attribute mapping for interoperability from IFC to CityGML. For this, we analyzed the interoperability issue including the neutral information model structure such as IFC, CityGML. To solve the interoperability problem between IFC and CityGML model which is the neutral GIS format, We proposed the meta data including the mapping rule. The meta data is consists of the connection information between BIM and GIS model, the mapping rule based on the perspective of the use-case, the operator and the attribute. By using this, XML to represent the meta data is defined and the information mapping system is developed.

Analytical Models for the Prediction of the Flexural Behavior for Thermal Bridge Breaker Systems embedded in Reinforced Concrete Slabs (열교차단장치가 적용된 철근 콘크리트 슬래브의 휨거동 예측을 위한 해석모델)

  • Shin, Dong-Hyeon;Kim, Hyung-Joon
    • Journal of the Computational Structural Engineering Institute of Korea
    • /
    • v.28 no.3
    • /
    • pp.325-333
    • /
    • 2015
  • Recently, thermal bridge breaker systems(TBBSs) applicable to RC slab-wall connections have been increasingly studied and proposed. This study also aims at proposing an analytic model which is applicable to predicting the flexural behavior of TBBS embedded in slabs from the initial elastic stages, yield states to ultimate conditions. The analytic models are developed by considering strain compatibility, force equilibrium and the constitutive law obtained from material test results. To verify the accuracy of the proposed analytic model, the moment-curvature relationship and change of neutral axis according to the loading states are compared with those of experimental results. Based on the comparison, it is verified that the proposed analytic model provides well predict the flexural behavior of TBBS embedded in slabs.

Robust Facial Expression Recognition using PCA Representation (PCA 표상을 이용한 강인한 얼굴 표정 인식)

  • Shin Young-Suk
    • Korean Journal of Cognitive Science
    • /
    • v.16 no.4
    • /
    • pp.323-331
    • /
    • 2005
  • This paper proposes an improved system for recognizing facial expressions in various internal states that is illumination-invariant and without detectable rue such as a neutral expression. As a preprocessing to extract the facial expression information, a whitening step was applied. The whitening step indicates that the mean of the images is set to zero and the variances are equalized as unit variances, which reduces murk of the variability due to lightening. After the whitening step, we used the facial expression information based on principal component analysis(PCA) representation excluded the first 1 principle component. Therefore, it is possible to extract the features in the lariat expression images without detectable cue of neutral expression from the experimental results, we ran also implement the various and natural facial expression recognition because we perform the facial expression recognition based on dimension model of internal states on the images selected randomly in the various facial expression images corresponding to 83 internal emotional states.

  • PDF

Optimum Slab-Lifting Positions for Precast Concrete Pavement Construction (프리캐스트 콘크리트 포장 시공 시 최적 슬래브 리프팅 위치)

  • Kim, Seong-Min;Cho, Byoung-Hooi;Han, Seung-Hwan
    • International Journal of Highway Engineering
    • /
    • v.9 no.2 s.32
    • /
    • pp.27-37
    • /
    • 2007
  • This research was conducted to determine the optimum lifting positions on precast concrete slabs for precast concrete pavement construction, based on the analysis of concrete stress distribution under various lifting conditions. To analyze stresses in concrete slabs, the finite element method was implemented and a numerical model of the precast slab that was going to be used in the experimental construction was developed. Changes in the stress distribution due to the lifting angle were investigated because slab lifting is not always performed in the perpendicular direction to the slab surface. In addition, the effect of the lifting level, the distance between the neutral axis of the slab and the lifting point, on the stress distribution was investigated since the lifting point is not always at the neutral axis of the slab. To consider the actual steel design of the precast slab, the effect of the reinforcement near the lifting point was also investigated. From this study, the optimum lifting positions of the precast slabs were determined according to the lifting angle and level, and the results were compared with the lifting positions used in the PCI standards.

  • PDF

Simplified Moment-Curvature Relationship Model of Reinforced Concrete Columns Considering Confinement Effect (구속효과를 고려한 철근 콘크리트 기둥의 모멘트-곡률 관계 단순모델)

  • Kwak, Min-Kyoung;Yang, Keun-Hyeok
    • Journal of the Korea Concrete Institute
    • /
    • v.28 no.3
    • /
    • pp.279-288
    • /
    • 2016
  • The present study simplified the moment-curvature relationship to straightforwardly determine the flexural behavior of reinforced concrete (RC) columns. For the idealized column section, moments and neutral axis depths at different stages(first flexural crack, yielding of tensile reinforcing bar, maximum strength, and 80% of the maximum strength at the descending branch) were derived on the basis of the equilibrium condition of forces and compatibility condition. Concrete strains at the extreme compression fiber beyond the maximum strength were determined using the stress-strain relationship of confined concrete, proposed by Kim et al. The lateral load-displacement curves converted from the simplified moment-curvature relationship of columns are well consistent with test results obtained from column specimens under various parameters. The moments and the corresponding neutral axis depth at different stages were formulated as a function of longitudinal reinforcement and transverse reinforcement indices and/or applied axial load index. Overall, curvature ductility of columns was significantly affected by the axial load level as well as concrete compressive strength and the amount of longitudinal and transverse reinforcing bars.

2-Parameter High Frequency Combustion Instability Model (2-파라메타 모델에 의한 고주파 연소불안정 해석)

  • 조용호;윤웅섭
    • Journal of the Korean Society of Propulsion Engineers
    • /
    • v.1 no.2
    • /
    • pp.74-83
    • /
    • 1997
  • The definition of burning admittance and conventional n-$\tau$ stability rating technique are combined to investigate the high frequency combustion instabilities inside the cylindrical combustion chamber. Perturbed flow variables are written as the sum of fluctuating and time-averaged mean quantities on the assumption that the terms of the order higher than unity are sufficiently small, hence linearized governing equations could be formulated. Chamber admittances up and downstream of the flame front calculated with appropriate boundary conditions result in the burning admittance and corresponding n-$\tau$ neutral stability curve. Configurational and operational design factors are tested to detect the unstable wave-induced LOX-RP1 combustion instabilities. Operational design factors, e.g. pressure or O/F ratio, appear less influential to drive high frequency instability while the location of the flame front and configurational factors enhance or deteriorate the stabilities strongly. Conclusively, LOX-RP1 combustion inside the cylindrical combustion chamber is apt to be unstable against long residence time and shortened chamber length.

  • PDF

Development of a Model for Analylzing and Evaluating the Suitability of Locations for Cooling Center Considering Local Characteristics (지역 특성을 고려한 무더위쉼터의 입지특성 분석 및 평가 모델 개발)

  • Jieun Ryu;Chanjong Bu;Kyungil Lee;Kyeong Doo Cho
    • Journal of Environmental Impact Assessment
    • /
    • v.33 no.4
    • /
    • pp.143-154
    • /
    • 2024
  • Heat waves caused by climate change are rapidly increasing health damage to vulnerable groups, and to prevent this, the national, regional, and local governments are establishing climate crisis adaptation policy. A representative climate crisis adaptation policy to reduce heat wave damage is to expand the number of cooling centers. Because it is highly effective in a short period of time, most metropolitan local governments, except Jeonbuk, include the project as an adaptation policy. However, the criteria for selecting a cooling centers are different depending on the budget and non-budget, so the utilization rate and effectiveness of the cooling centers are all different. Therefore, in this study, we developed logistic regression models that can predict and evaluate areas with a high probability of expanding cooling centers in order to implement adaptation policy in local governments. In Incheon Metropolitan City, which consists of various heat wave-vulnerable environments due to the coexistence of the old city and the new city, a logistic model was developed to predict areas where heat waves can be cooling centered by dividing it into Ganghwa·Ongjin-gun and other regions, taking into account socioeconomic and environmental differences. As a result of the study, the statistical model for the Ganghwa·Ogjin-gun region showed that the higher the ground surface temperature and the more and more the number of elderly people over 65 years old, the higher the possibility of location of cooling centers, and the prediction accuracy was about 80.93%. The developed logistic regression model can predict and evaluate areas with a high potential as cooling centers by considering regional environmental and social characteristics, and is expected to be used for priority selection and management when designating additional cooling centers in the future.