• Journal of Korea Water Resources Association
• /
• v.37 no.8
• /
• pp.663-674
• /
• 2004

In this study, the principle of water allocation is proposed based on efficiency, equity, sustainability. Also weighting factors are estimated with sectoral factors and regional factors. The former represents relative weights among water use and the latter represents physical characteristics of water demand places. The AHP(Analytic Hierarchy Process) is applied to estimate the sectoral factors, and compounded regional-characteristic factors and regional-scale factors, which reflects socioeconomic statistics for the regional factors. By applying these weighting factors, water allocation rules for dam is developed and applied to Andong dam which supplies water to parts of Busan Shi, Daegu Shi and Goryeong Gun in a water-deficit situation. As a result, it is estimated that Water allocation by priorities distributes the entire water shortage to the lowest rank of water sectors or regions, while water allocation by relative weighting factors disperse all the burdens of water shortage to all sectors and regions.

• Journal of Environmental Impact Assessment
• /
• v.26 no.3
• /
• pp.171-180
• /
• 2017

The vulnerability analysis of climate change driven disaster has been used as institutional framework for the urban policies of disaster prevention since 2012. However, some problems have occurred due to the structure of vulnerability analysis, such as overweighted variables and duplicated application of variables of similar meaning. The goal of this study is to examine the differences of results between the method of current guideline and the method of weight equalization. For this, we examines the current structural framework of the vulnerability analysis, and performs empirical analysis. As a result, the extent and magnitude of vulnerability showed different spatial patterns depending on the weighting methods. Standardized weighting method relatively represented wider vulnerable areas compared to the pre-existing method which follows the current instruction manual. To apply the results of vulnerability analysis to urban planning process for disaster prevention, this study suggests that the reliability of the results should be ensured by improving analytical framework and detailed review of the results.

• The Journal of the Korea Contents Association
• /
• v.14 no.4
• /
• pp.160-167
• /
• 2014

The purpose of this paper was to develop the key performance indicators of regional co-training programs, using analytic hierarchy process method. Firstly, we pooled indicators by reviewing the literatures. The researcher checked the properness and representative of them and then, the training professionals reviewed them again. Finally, we finalized the indicators, depending on the indicator domains such as MOU (Memorandum of Agreement) company (the training participation rate of MOU companies, training participation of employees, and etc.), training performance (the completion of the program, employment rate, and etc.), cost-effectiveness (training cost per capita), satisfaction (satisfaction rate of MOU companies and training participants), and regional co-training enhancement (the operation of the committee, the will for training of the regional municipal government, and etc.). The weight size varied by the regional training needs, satisfaction of MOU companies, and regional committees of human resource development. We also suggested the continuous revision of the key performance indicators for regional co-training.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.265-268
• /
• 2007

다기준 의사결정 문제에서 요인간의 가중치 계산과 계산된 요인의 평가값 종합화는 매우 중요하다. 본 연구는 다기준 의사결정 문제에 있어서 의사결정자의 의사전략 결합기법을 도출하고 다기준의사결정 문제로 적용하였다. 복잡한 환경에서 의사결정을 할 때 발생되는 모호함을 해결하기 위해 주관적 의견을 결합한 퍼지지합 이론을, 다기준 문제의 요인을 퍼지값으로 계층화하기 위해 계층분석법을 적용하였다. 또한, 의사결정자의 의사전략을 결합하기 위해 순위 가중치평균법을 이용하였다. 순위가 있는 가중치 평균방법은 퍼지집합의 orness 특성을 이용하여 의사결정자의 주관적 의지를 반영할 수 있는 기법으로, 순위가중치평균(OWA) 연산자에 따른 낙관적 혹은 비관적인 정도에 따라 주관적인 의도를 반영할 수 있는 방법이다. 다기준의사결정 문제의 적용사례로서 해상교통안전을 위한 대기정박지의 위치분석 문제를 본 연구에서 제시한 방법에 따라 적용하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2011.01a
• /
• pp.39-42
• /
• 2011

본 논문에서는 영상들의 특징들을 추출하여 특징 값들의 비교를 통하여 질의 영상의 유사 영상을 검색하는 방법을 제안한다. 제안하는 방법은 입력 영상들의 색상 히스토그램으로 색상 특징 값들을 추출하고 질감 정보인 에지 정보와 이웃화소간의 공간 관계를 분석하여 질감 특징 값들을 추출하여 저장한 후 질의 이미지의 색상과 질감 특징들을 구하여 비교를 통하여 유사도를 분석하고 결과 영상을 보여준다. 또한 색상과 질감을 혼합하여 사용할 때 적응적으로 가중치를 부여함으로써 가중치가 적합하지 않아 발생하는 오 검출될 현상을 피할 수 있게 되었다. 실험을 통하여 기존의 방법과의 성능을 비교분석하였고 본 방법의 우수성을 입증하였다.

• Journal of Internet Computing and Services
• /
• v.17 no.1
• /
• pp.65-71
• /
• 2016

In this paper, we propose a novel analysis method and its algorithms to perform the betweenness centrality measurements on a valued and directed workflow-supported social network. The conventional workflow-supported social network model is to represent the existences of task transferring relationships among their performers by using a binary social network. However, it is necessary to consider not only the existences of task transferring relationships but also their quantities and directions in order to obtain much more effective and sophisticated analysis results. In conclusion, this paper newly defines a concept of valued and directed workflow-supported social network, and its betweenness centrality analysis method and algorithms. Especially, to verify the proposed method and algorithms, we try to apply the conventional method and the proposed method to an example workflow model respectively, and compare their betweenness centrality analysis results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10c
• /
• pp.409-411
• /
• 2002

본 논문에서는 스트리밍 미디어 캐슁 서버의 효율적인 캐슁 구조를 위하여 참조 횟수, 참조량, 참조 시간 둥의 정량적인 인자들과 사용자 요구 주기를 적용하여 최근 참조 경향에 높은 가중치를 부여함으로써 변화하는 콘텐츠 선호 경향에 빠르게 적응하는 가중치 기반의 캐쉬 교체 정책을 제안한다. 성능 분석은 시뮬레이션 환경 구축을 통해 실험하였으며 LRU, LFU와 SEG 캐쉬 정책과 비교 분석하여 향상된 결과를 보였다.

• Journal of Digital Convergence
• /
• v.14 no.10
• /
• pp.167-172
• /
• 2016

Commercial analysis has been used for a lot of reasons such as improving the distribution structure, operational cost savings. However, existing commercial assay system is mostly flow through the analysis of the population and the environment around the store. Therefore, in this paper, we design a path weighted data analysis system for efficient product arrangement. It is not applying the analysis method of the existing commercial sales center but the data analysis system applying the weight to the form of customers' paths in a store. In conjunction with the existing POS system, it provides effective product placement in department stores, such as supermarkets, convenience stores. To this end, a system that maximizes efficiency of product placement in stores, receiving path data, converting them into statistics data that apply the weight, and printing out them into a visualized monitering system for the manager to see was designed.

• Proceedings of the KSRS Conference
• /
• 2008.03a
• /
• pp.237-240
• /
• 2008

본 연구의 목적은 인공신경망 기법을 이용하여 2006년 전국을 대상으로 구축된 국토환경성평가지도의 환경 생태적항목에 대한 각 항목별 가중치를 결정하는 것이다. 기본 분석 도구로 지리정보시스템(GIS)가 사용되었다. 국토환경성평가지도의 환경 생태적항목은 다양성(생태자연도), 자연성(임상도, 녹지자연도, 생태자연도), 풍부도(생태계변화관찰 지역도), 희귀성(생태자연도), 허약성(수치지형도, 토지피복도), 군집구조의 안정성(임상도)등이 활용되어 구축되었다. 본 연구는 기 구축되어 사용되고 있는 국토환경성평가 지도의 환경 생태적항목을 공간 데이터베이스를 이용하고, 인공신경망 기법을 적용하여 각 평가항목간의 상대적 가중치를 구하였다. 인공신경망의 훈련 지역은 환경 생태적항목중 환경성이 높은 1등급 지역 및 환경성이 낮은 5등급 지역을 추출하였다. 그 결과 50번의 가중치를 산정하였을 경우 허약성이 다른 항목들에 비해 1.58배 정도 높은 상대적 가중치를 나타냈다. 이러한 가중치는 국토환경성평가지도 환경 생태적 항목의 취약성도를 작성하는데 활용될 수 있다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.9
• /
• pp.77-81
• /
• 2016

Logistic regression (LR) has been widely used for predicting the relationships among variables in various fields. We propose a new logistic regression model with a fine-grained weighting method, called value weighted logistic regression, by assigning different weights to each feature value. A gradient approach is utilized to obtain the optimal weights of feature values. We conduct experiments on several data sets and the experimental results show that the proposed method shows meaningful improvement in prediction accuracy.