• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2005년도 춘계공동학술대회 발표논문
• /
• pp.614-620
• /
• 2005

Ensemble에서 feature selection은 각 classifier의 학습할 데이터의 변수를 다르게 하여 diversity를 높이며, 이것은 일반적인 성능향상을 가져온다. Feature selection을 할 때 쓰는 방법 중의 하나가 Genetic Algorithm (GA)이며, GA-SVM은 GA를 기본으로 한 wrapper based feature selection mechanism으로 response model과 keystroke dynamics identity verification model을 만들 때 좋은 성능을 보였다. 하지만 population 안의 후보들간의 diversity를 보장해주지 못한다는 단점 때문에 classifier들의 accuracy와 diversity의 균형을 맞추기 위한 heuristic parameter setting이 존재하며 이를 조정해야만 하였다. 우리는 GA-SVM 알고리즘을 바탕으로, population안 후보들의 fitness를 측정할 때 accuracy와 diversity 둘 다 고려하는 fitness function을 도입하여 추가적인 classifier 선택 작업을 제거하면서 성능을 유지시키는 방안을 연구하였으며 결과적으로 알고리즘의 복잡성을 줄이면서도 모델의 성능을 유지시켰다.

• 한국경영과학회지
• /
• 제28권1호
• /
• pp.63-77
• /
• 2003

Group buying is one of the most popular transaction patterns on the internet at least in Korea. Nevertheless, it is hard to find academic researches for it in view of consumer behavior. In this paper, we analyze factors which determine consumer's attitude toward and intention of participation in Internet group buying by comparison of Internet individual one. For this purpose, we propose “lowering price”, “decreasing risk” and “reducing transaction cost” as relative advantages and "lack of product assortment" and "delay of time" as relative disadvantages over individual buying on the Internet. For empirical test, Internet users who have some experiences of individual buying but not group ones on the Internet are surveyed and analyzed. In result, a satisfying model fitness for structural equation model is derived and most hypotheses except the relationship between "decreasing risk" and "attitude toward Internet group buying" are accepted. Our results provide not only academic contribution by suggestion of a research framework but also practical insight by discussion of diverse features in Internet group buying.verse features in Internet group buying.

• 한국농공학회지
• /
• 제39권6호
• /
• pp.54-66
• /
• 1997

The objectives of this study is to introduce and apply neural network theory to real hydrologic systems for stochastic nonlinear predicting of daily runoff discharge in the river catchment. Back propagation algorithm of neural network model is applied for the estimation of daily stochastic runoff discharge using historical daily rainfall and observed runoff discharge. For the fitness and efficiency analysis of models, the statistical analysis is carried out between observed discharge and predicted discharge in the chosen runoff periods. As the result of statistical analysis, method 3 which has much processing elements of input layer is more prominent model than other models(method 1, method 2) in this study.Therefore, on the basis of this study, further research activities are needed for the development of neural network algorithm for the flood prediction including real-time forecasting and for the optimal operation system of dams and so forth.

• 제어로봇시스템학회논문지
• /
• 제10권9호
• /
• pp.755-762
• /
• 2004

In this research, EOG(Electrooculogram) signal was analyzed to predict the subject's intention using a fuzzy classifier. The fuzzy classifier is built automatically using the EOG data and evolutionary algorithms. An assistant robot manipulator in redundant configuration has been developed, which operates according to the EOG signal classification results. For automatic fuzzy model construction without any experts' knowledge, an evolutionary algorithm with the new representation scheme, design of adequate fitness function and evolutionary operators, is proposed. The proposed evolutionary algorithm can optimize the number of fuzzy rules, the number of fuzzy membership functions, parameter values for the each membership functions, and parameter values for the consequent parts. It is shown that the fuzzy classifier built by the proposed algorithm can classify the EOG data efficiently. Intelligent motion planner that consists of several neural networks are used for control of robot manipulator based upon EOG classification results.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 1998년도 추계종합학술대회
• /
• pp.146-150
• /
• 1998

본 논문에서는 유전자 알고리즘을 이용하는 새로운 시스템 식별 방식을 제안한다. 제안 한 방식은 ARMAX 모델을 이용하여 비선형 시스템을 파라미터 벡터와 측정 벡터로 나누고, 파라미터 벡터를 유전자 알고리즘을 이용하여 최적의 값을 구하여 ARMAX 모델의 파라미터를 조정한다. 기존의 Narendra의 4가지 식별 모델을 대상으로 시뮬레이션하여 제안한 식별 방식의 유용성을 확인하였다.

• 한국생산제조학회지
• /
• 제22권2호
• /
• pp.241-247
• /
• 2013

The doctor blade is equipped in a toner cartridge and is a device to maintain the uniform thickness of a toner by controlling the pressure on the developing roller. The contact pressure between the developing roller and the doctor blade is one of the significant factors for image quality and durability of toner cartridge. The purpose of this study is to develop approximation model in order to minimize the time and cost which are needed much required in making optimal design of the doctor blade. Central composite design was used for the design of experiment and response surface design was used for approximation. The data for contact pressure were acquired through finite element analysis and data of image density and toner weight were acquired through experiment. The approximation model developed in this study has presented very high fitness.

• 산업경영시스템학회지
• /
• 제23권58호
• /
• pp.119-128
• /
• 2000

The purpose of this study was to verify the determinants of person-organization(P-O) fit and its effects on job satisfaction and turnover intention. As a result of empirical survey, value congruence, job choice decision and similar occupation are confirmed as factors that affect on P-O Fit. And the effect of P-O fit on job satisfaction and turnover intention was significant. This findings indicated that P-O fit model can be used for organization to improve employee performance. Some limitations of this study and strategic application of P-O fit model for employee selection and retention were also discussed.

• 한국식품조리과학회지
• /
• 제28권3호
• /
• pp.285-298
• /
• 2012

This study was undertaken to understand the interrelationships among customer's perceptions of service encounters, customer value, and behavioral intent in a family restaurant. Based on 331 samples obtained from empirical research, we reviewed the reliability and fitness of a research model and verified three hypotheses using a structural equation model. The results showed that physical environment (${\beta}$ = .418) and a customer interaction with the service provider (${\beta}$ = .265) had a significant effect on the customer's hedonic value. Additionally, physical environment (${\beta}$ = .126) and customer interaction with service providers (${\beta}$ = .264) had a significant effect on customer utilitarian value. Customer's hedonic (${\beta}$ = .538) and utilitarian value (${\beta}$ = .382) triggered by service encounters had a significant effect on their behavioral intent. Limitations and future research directions are discussed.

• 수산해양교육연구
• /
• 제19권1호
• /
• pp.8-18
• /
• 2007

In order to define their relative influence, this study analyzed causal relationships between school adjustment of middle school students and such related variables as locus of control and stress, including clarification of other significant variables affecting school adjustment. Participants were a sample of 331 students from 3 middle schools in Busan. Correlation, analysis of covariance structure modeling, and model-fitness tests were applied to the data. Results indicated that stress on the school, home, social relationships, and self has the more significant influence on the school adjustment, and the locus of control has relatively low influence.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1996년도 추계학술대회
• /
• pp.70-73
• /
• 1996

Many researchers had considered biological systems as linear systems. In many cases of biological systems, the phenomena that show the regular and periodic dynamics are considered the normal state. However, some clinical experiments reported, in some cases, the periodic signals represented the abnormal state. We assume that signals from human body system are generated from deterministic, intrinsic mechanisms and can be represented a simple equation that show nonlinear dynamics dependent on control parameters. The objective of our study is to model a nonlinear dynamics correctly from the nonlinear time series using the genetic programming method; to find a simple equation of nonlinear dynamics using collected time series and its nonlinear characteristics. We applied genetic programming to model RR interval of ECG that shows chaotic phenomena. We used 4 statistic measures and 2 fractal measures to estimate fitness of each chromosome, and could obtain good solutions of which chaotic features are similar.