• 전자공학회논문지CI
• /
• 제41권1호
• /
• pp.1-8
• /
• 2004

인터넷의 발달과 데이터베이스의 구축이 보편화됨에 따라 막대한 양의 데이터 속에서 의사 결정에 필요한 지식을 찾아내는 작업은 결코 쉬운 일이 아니다 본 논문에서는 대규모 데이터의 효율적인 분석을 위하여 지식의 탐사 이전에 데이터에 대한 축소 작업을 수행하기 위한 효과적인 차원 축소 전략에 의한 패턴분류 기법을 제안한다. 이를 위해 본 논문에서는 통계적학습 모형인 Support Vector Machine의 VC-dimension에 기반한 RBF 신경망 모형을 제안한다. 기존의 RBF 신경망 모형은 주로 퍼셉트론 모형의 전처리 작업만을 수행하지만 제안하는 신경망 모형은 VD-dimension과 연계한 독자적으로 데이터를 분석할 수 있는 능력을 갖춘 모형을 구축하고 이를 바탕으로 개체들을 정확한 레이블로 분류한다. 기계 학습 데이터를 이용하여 본 논문에서 제안하는 모형의 성능을 비교 평가한 결과 기존의 여러 분류 알고리즘에 비해 우수한 성능을 보임이 실험을 통해 확인되었다.

• 전자공학회논문지CI
• /
• 제48권5호
• /
• pp.42-48
• /
• 2011

에이전트는 특정 목적을 위해 행동을 하는데 이것은 자율지능형 가상군(Computer Generated Forces, CGF)의 공통된 요소이다. 목적을 달성하기 위해 지정된 스크립트를 따라 행위를 하거나 업무 수행의 계획을 세우는 것을 기본적인 에이전트의 지능이라 볼 수 있는데 이보다 더 발전된 지능 에이전트는 계획을 세우는 것뿐만 아니라 계획했던 수행이 어려울 때 계획을 다시 수정하거나 새로운 계획을 적응적으로 만들어내는 것이다. 계획을 수행 할 때 에이전트가 목적을 위한 적응적 행동을 하려면 목표를 달성할 가능성이 적어질 때 스스로 계획을 수정하고 이러한 방식으로 수정되는 방법을 계속적으로 학습하여 차후 같은 경우에는 학습이 반영된 더 좋은 계획 및 전술을 반영하도록 해야 한다. 즉, 목표와 현재의 상태를 실시간으로 분석하고 측정하여 목표 달성도를 정량적으로 계산하고 측정값이 임계값보다 적으면 수정된 계획을 선택하도록 하는 것이다. 본 논문에서는 위와 같이 에이전트가 목표 달성 가능성이 적어질 때 적응적으로 계획을 새롭게 수정하여 적용하는 방법을 연구한 목적 지향적 행위계획 방법을 제안한다.

• 재활간호학회지
• /
• 제5권2호
• /
• pp.205-216
• /
• 2002

This quasi-experimental study was done to test the effects of meta-memory enhancing program for elderly women. Data were collected 12 to 30, August 2002 from 34elderly women over 65 years living in Busan city. Subjects were 15 of experimental group and 19 of control group. The metamemory enhancing program was developed by five sessions composing of 1.5-2.0 hours one session. In experiment group, this program was performed for three weeks, twice per week. The degrees of four memory performance tasks were measured using instrument of Elderly Verbal Learning Test(Choi Kyung Mi, 1988) and Face Recognition Instrument(Min Hye Sook, 1999) and the metamemory were measured using MIA questionnaire(Dixon et al., 1988). Research results are as following. 1. After participating in five times memory training programs, experimental group has the significant increase of metamemory in comparison with control group.(t=59.58, p< 0.0001). In particular, the concepts of strategy(t=20.44, p< 0.0001), achievement (t=21.94, p< 0.0001), and locus degree (t=59.58, p< 0.0001) among sub-concepts of the metamemory are increasing significantly. 2. After participating in five time memory training programs, the degree of immediate word recall(t=17.25, p< 0.0001) and face recognition(t=16.69, p< 0.0001) among four memory tasks in experimental group are increasing significantly compared with those measures of control group. Considering this results, this metamemory enhancing program was found as an effective nursing program for metamemory improvement of elderly women's memory.

• 한국의상디자인학회지
• /
• 제16권4호
• /
• pp.79-98
• /
• 2014

As men are interested in fashion and beauty and invest for himselves nowadays, the scale of men's cosmetic businesses is showing steadily tendency. The purposes of this were to offer data to establish a marketing strategies learning follows: The findings of this study were as follows: First, according to the survey on men's consideration of their look, it showed their high interest in it. 29% respondents carry cosmetics in their hands normally. Second, Men's level of skin management knowledge is quite high according the survey. Third, The largest group of respondents are obtaining information about cosmetics through mass media. therefore, there will be necessity for cosmetic firms to aggressively pursue positive publicity strategy various of mass media. Fourth, Regarding purchasing behavior, 53.2% buy cosmetics in person, who outnumbered the others who don't. In the case of men who don't buy cosmetics by themselves, spouses or girl friends purchase cosmetics on behalf of them. Fifth, as a result of examine the demand level for men's cosmetics, the men, who answered that it is appropriate a lotion and toner is priced between 10,000 won and 30,000 won were the largest with 51.1%. Men who answered that beside the basic cosmetics such as toner and lotion that use, they will purchase sunscreen in the future. Men sho recognize the most important task needed to be improved in men's cosmetics is higher quality, with 25.9%.

• 한국데이타베이스학회:학술대회논문집
• /
• 한국데이타베이스학회 1999년도 국제컨퍼런스 디지털컨텐츠 활용을 통한 지식경영의 확산
• /
• pp.183-189
• /
• 1999

오늘의 기업에게 정보기술이란 필수요소이다. 정보기술을 효과적으로 활용하기 위해서는 IT 인프라가 체계적으로 구축되어 있어야 한다. 해당 조직에 적합한 정보기술 기반구조를 설계하고 이의 도입을 위한 구체적인 계획을 수립하기 위해서는 체계적이고 효과적인 정보계획 수립(Information System Planning: ISP)이 필요하다. 그러나 정보계획수립 프로젝트의 상당수가 실패로 그치고 있다. 특히 정보기술의 지속적인 변화 때문에 수립한 정보기술 기반구조 계획안이 실제 구현되지 못하고 계획으로만 남는 경향이 있다. 이러한 ISP의 어려움을 해결하기 위해서는 정보기술 참조모델(reference model)을 적극적으로 활용할 필요가 있다. 즉, 조직의 정보시스템에 공통적으로 적용할 수 있는 IT 인프라나 표준 아키텍쳐를 바탕으로 정보계획수립을 수행해 나가는 방식이 필요하다. 이와 같은 참조모델 기반의 정보계획 수립은 새로운 아키텍쳐를 추출하고 표준화를 이룸으로써 프로젝트의 생산성을 높일 수 있다는 장점을 가지고 있다. 기존의 ISP 연구는 ISP의 필요성, 과정, 성공요인 등에 국한되어 왔으며, 방법론에 대한 연구는 미비한 편이다. 최근들어 ISP의 체계적인 분류나 참조모델 기반 계획수립의 필요성이 제기되었다. 그러나 아직까지 이와같은 접근에서 참조 모델을 어떻게 구축하고 활용해 나갈 것인가에 대한 연구는 부족한 실정이다. 따라서 본 논문에서는 참조모델을 구축하기 위한 다양한 접근방법과 각각의 특징을 제시한다. 나아가서 해당 조직의 상황이나 요구수준에 따라 적합한 접근방법을 선택할 수 있게 해주는 방안을 제시한다.타냈으며, 평가결과에 대해 여러 가지 방법으로 분석하였다. 첫째, 동종제품간 평가분석을 통하여 각각의 제품을 비교하였으며, 둘째 소프트웨어 종류별 평가로 제품을 응용소프트웨어, 응용개발도구, 시스템 소프트웨어로 분류하여 평균값으로 비교하였다. 셋째, 국내외 제품별 평가분석으로 전체 제품을 국내제품과 국외제품으로 분류하여 비교하였으며, 마지막으로 총괄분석을 통해 가중치를 적용하여 전 제품의 점수를 비교하였다. 여기에서는 각 제품의 평균점수에 대한 차이를 95%의 유의수준으로 T-Test를 실시하였다.uted to the society, and what the socioeconomic impacts are resulted from the program. It would be useful for the means of (ⅰ) fulfillment of public accountability to legitimate the program and to reveal the expenditure of pubic fund, and (ⅱ) managemental and strategical learning to give information necessary to improve the making. program and policy decision making, The objectives of the study are to develop the methodology of modeling the socioeconomic evaluation, and build up the practical socioeconomic evaluation mod

• Journal of Information Technology Applications and Management
• /
• 제12권2호
• /
• pp.79-106
• /
• 2005

This study, by adopting case study methodology, is focused on examining the present state and analyzing the cause of the digital divide, and suggesting policies for bridging the divide, specifically in view of SMEs. We have taken cases of manufacturing companies, visiting and interviewing 18 SMEs in 10 APEC member economies which show sharp difference in usage of ICT. In order to analyze the digital gap among SMEs, we used 5 variables that are composed of computer hardware, computer software, Internet, readiness of ICT, and performance of ICT adoption, while categorizing the cases into low and high tier based on the national ICT index. From a computer hardware perspective, the high tier (0.66) has almost double the number of PC’s per employee, compared with the low tiers (0.34). This gap can be explained by financial availability of low income and high tariff in the developing economies. In the computer software perspective, the SMEs in the low tier had some restrictive use of computer applications such as financial and accounting management and document management, while those in the high tier enjoyed more diversity in the use of applications such as inventory management, sales management, financial and accounting management, procurement management, CRM, and ERP. In view of the readiness of ICT, the difference in ICT infrastructure and financial status between the low and high tier was far wider than any other variables. As a result of ICT adoption, SMEs benefited in view of learning and growth, internal business processes, customer service, and financial affairs. To effectively bridge the digital divide between the low and high tier, actions such as setting up a secondary market of used computers among cooperating developed and developing countries, developing and diffusing good business applications, and building speedy, low-cost telecommunication infrastructures should be taken.

• 정보처리학회논문지B
• /
• 제12B권4호
• /
• pp.429-436
• /
• 2005

본 논문에서는 유전자 알고리즘의 일반적인 문제점인 과도한 저장공간의 소모와 탐색의 비효율성을 줄이기 위해 PBIL을 이용한 단순한 스테레오 정합 기법을 제안한다. PBIL은 확률벡터에 기반해서 통계적 탐색과 경쟁학습을 이용하는 변종 유전자 알고리즘이며 확률벡터의 사용으로 인해 직렬 및 병렬 유전자 알고리즘군에 비해 단순한 구조를 가진다. 본 논문에서는 이 PBIL을 스테레오 정합 환경에 맞게 변형 및 단순화시켜 정합 알고리즘을 개발한다. 높은 적응성을 갖는 염색체는 생존 확률 또한 높다는 진화 법칙을 보존하면서 유전자 풀, 염색체 교차 및 유전자 돌연변이를 제거할 수 있으며 그 결과 저장공간을 줄이고 정합 규칙을 간소화하여 계산 비용을 감소시킬 수 있다. 추가적으로 다해상도 정합 기법처럼 넓은 영역의 변이 일관성을 획득하기 위해 변이 연속성에 대한 이웃들의 거리를 제어하는 방식을 추가하여 고정된 작은 정합창을 사용하면서 안정된 결과를 얻을 수 있게 한다. 마지막으로 단순한 시스템에 적용될 수 있게 하기 위해서 확률벡터를 사용하지 않는 제안한 알고리즘의 소형 대안 기법을 제시한다.

• 전기학회논문지
• /
• 제60권4호
• /
• pp.862-870
• /
• 2011

In this paper, we introduce a new topology of Radial Basis Function-based Polynomial Neural Networks (RPNN) that is based on a genetically optimized multi-layer perceptron with Radial Polynomial Neurons (RPNs). This study offers a comprehensive design methodology involving mechanisms of optimization algorithms, especially Fuzzy C-Means (FCM) clustering method and Particle Swarm Optimization (PSO) algorithms. In contrast to the typical architectures encountered in Polynomial Neural Networks (PNNs), our main objective is to develop a design strategy of RPNNs as follows : (a) The architecture of the proposed network consists of Radial Polynomial Neurons (RPNs). In here, the RPN is fully reflective of the structure encountered in numeric data which are granulated with the aid of Fuzzy C-Means (FCM) clustering method. The RPN dwells on the concepts of a collection of radial basis function and the function-based nonlinear (polynomial) processing. (b) The PSO-based design procedure being applied at each layer of RPNN leads to the selection of preferred nodes of the network (RPNs) whose local characteristics (such as the number of input variables, a collection of the specific subset of input variables, the order of the polynomial, and the number of clusters as well as a fuzzification coefficient in the FCM clustering) can be easily adjusted. The performance of the RPNN is quantified through the experimentation where we use a number of modeling benchmarks - NOx emission process data of gas turbine power plant and learning machine data(Automobile Miles Per Gallon Data) already experimented with in fuzzy or neurofuzzy modeling. A comparative analysis reveals that the proposed RPNN exhibits higher accuracy and superb predictive capability in comparison to some previous models available in the literature.

• KIEAE Journal
• /
• 제17권5호
• /
• pp.69-76
• /
• 2017

Purpose: This study aimed at developing an Artificial Neural Network (ANN) model for predicting the amount of cooling energy consumption of the variable refrigerant flow (VRF) cooling system by the different set-points of the control variables, such as supply air temperature of air handling unit (AHU), condenser fluid temperature, condenser fluid pressure, and refrigerant evaporation temperature. Applying the predicted results for the different set-points, the control algorithm, which embedded the ANN model, will determine the most energy efficient control strategy. Method: The ANN model was developed and tested its prediction accuracy by using matrix laboratory (MATLAB) and its neural network toolbox. The field data sets were collected for the model training and performance evaluation. For completing the prediction model, three major steps were conducted - i) initial model development including input variable selection, ii) model optimization, and iii) performance evaluation. Result: Eight meaningful input variables were selected in the initial model development such as outdoor temperature, outdoor humidity, indoor temperature, cooling load of the previous cycle, supply air temperature of AHU, condenser fluid temperature, condenser fluid pressure, and refrigerant evaporation temperature. The initial model was optimized to have 2 hidden layers with 15 hidden neurons each, 0.3 learning rate, and 0.3 momentum. The optimized model proved its prediction accuracy with stable prediction results.

• 한국학교ㆍ지역보건교육학회지
• /
• 제6권
• /
• pp.1-16
• /
• 2005

This study was conducted to the middle and high school students in some Seoul and Kyunki areas to identify the smoking behavior characteristics among adolescences. A self-administered survey was conducted to the 2nd grad students in 4 middle schools and 6 high schools and the survey Questionnaire included general characteristics, smoking and drug use history, the close people's smoking and drug use, smoking and drug abuse prevention education, smoking intention, and smoking attitude. A total of 2,452 youths finished the survey (1,182 middle school students and 1,270 high school students). Current smoking students were 14.6%, the ex-smokers were 5.5%, and the never smokers were 85.4%. Majority of students smoked less than 5 bars of cigarettes and their first smoking experiences were related to their family members (siblings, parents, and relatives), friends, advertisement in order. Other GYTS countries reported the similar sources of the smoking start and friend was prior smoking start factor to the other sources. The students who wanted to Quit smoking were 6.7% and the students who ever had tried to Quit smoking were 9.1%. The major reasons of Quitting smoking were for their health and for their financial burden. Approximately 60% learned about smoking and drug abuse in their regular school classes, 8.4% were in the special school activities, and 7.9% were in the class closing time sometimes in order. The students who learned in any regular class were smaller in the high school students than in the middle school students. The learning experiences in school of other GYTS countries were similar to that of Korea. In conclusion, students' smoking was affected not only by the preventive activities in school but also by the close people's behaviors and care in this study; therefore, the active partnership between school and family must be a strong strategy for youth's smoking prevention.