• 종양간호연구
• /
• 제2권1호
• /
• pp.5-16
• /
• 2002

Purpose: This study was to develop a stress-adaptation model for family caregivers of cancer patients that could provide the basis of planning nursing intervention. Method: A hypothetical model was developed using the family adaptation model proposed by Haley et al. (1987). In the literature, the stressor was identified as patient's characteristics, caregiver's characteristics, duration of illness, and family life events. It affected stress appraisal, family resources, family coping and finally caregiver's adaptation. In this model, 18 paths were constructed. Data were collected from 241 caregivers, whose family members were in treatment between June and August 2000, at 3 university hospitals and were analyzed by SPSS and LISREL programs. Results: 1) The overall fitness indices of the hypothetical model were x 2=267.78 (P= .0), GFI= .92, AGFI= .87, NFI= .93, NNFI= .93, PNFI= .64, PGFI= .55, and RMR= .43. Ten of the eighteen paths proved to be significant. 2) To improve the model fitness, the hypothetical model was modified considering modification indices and the paths proved not significant. Final model excluded 3 paths demonstrated to be improved by x2=161.96 (P= .00), GFI= .95, AGFI= .91, NFI= .96, NNFI= .96, and RMR= .23. Twelve of fifteen paths proved to be significant. 3) Stress appraisal was influenced by disease related characteristics and duration of illness and was explained 22% of the variance. Family resources were influenced by stress appraisal and was explained 57% of variance. Family coping was influenced by disease related characteristics, caregiver's characteristics, duration of illness, family life event, and stress appraisal and was explained 57% of variance. Family caregiver adaptation was influenced by disease related characteristics, caregiver's characteristics, stress appraisal, and family coping and was explained 31% of variance. Twelve of fifteen paths were significant. Conclusion: Based on this study, to help family caregivers to adapt, individual intervention is necessary with consideration of disease related and caregiver's characteristics and duration of illness. The intervention should include efforts to raise the family resources and to identify positively the stress they encounter, and there is a need to establish an adaptation model that considers emotional aspects of family caregivers. Since there is a difference in emotional status depending on the disease stage, a study needs to be done to analyze the differences among the disease stages (diagnosis, treatment, recurrence, and terminal stages).

• 지능정보연구
• /
• 제20권4호
• /
• pp.121-139
• /
• 2014

기업의 부도 예측은 재무 및 회계 분야에서 매우 중요한 연구 주제이다. 기업의 부도로 인해 발생하는 비용이 매우 크기 때문에 부도 예측의 정확성은 금융기관으로서는 매우 중요한 일이다. 최근에는 여러 개의 모형을 결합하는 앙상블 모형을 부도 예측에 적용해 보려는 연구가 큰 관심을 끌고 있다. 앙상블 모형은 개별 모형보다 더 좋은 성과를 내기 위해 여러 개의 분류기를 결합하는 것이다. 이와 같은 앙상블 분류기는 분류기의 일반화 성능을 개선하는 데 매우 유용한 것으로 알려져 있다. 본 논문은 부도 예측 모형의 성과 개선에 관한 연구이다. 이를 위해 사례 선택(Instance Selection)을 활용한 배깅(Bagging) 모형을 제안하였다. 사례 선택은 원 데이터에서 가장 대표성 있고 관련성 높은 데이터를 선택하고 예측 모형에 악영향을 줄 수 있는 불필요한 데이터를 제거하는 것으로 이를 통해 예측 성과 개선도 기대할 수 있다. 배깅은 학습데이터에 변화를 줌으로써 기저 분류기들을 다양화시키는 앙상블 기법으로 단순하면서도 성과가 매우 좋은 것으로 알려져 있다. 사례 선택과 배깅은 각각 모형의 성과를 개선시킬 수 있는 잠재력이 있지만 이들 두 기법의 결합에 관한 연구는 아직까지 없는 것이 현실이다. 본 연구에서는 부도 예측 모형의 성과를 개선하기 위해 사례 선택과 배깅을 연결하는 새로운 모형을 제안하였다. 최적의 사례 선택을 위해 유전자 알고리즘이 사용되었으며, 이를 통해 최적의 사례 선택 조합을 찾고 이 결과를 배깅 앙상블 모형에 전달하여 새로운 형태의 배깅 앙상블 모형을 구성하게 된다. 본 연구에서 제안한 새로운 앙상블 모형의 성과를 검증하기 위해 ROC 커브, AUC, 예측정확도 등과 같은 성과지표를 사용해 다양한 모형과 비교 분석해 보았다. 실제 기업데이터를 사용해 실험한 결과 본 논문에서 제안한 새로운 형태의 모형이 가장 좋은 성과를 보임을 알 수 있었다.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1997년도 춘계학술대회논문집
• /
• pp.155-158
• /
• 1997

The investigation deals with of a intermediate process condition hving a bolt-shaped final product where it is required to extend tool-life in forging. In this study, optimization of the design variables is conducted by a genetic algorithm, where the fitness values are evaluated on the basis of FEM analysis model. The approach is applied to the determination of the intermediate process conditions which are optimal with regard to minimization of peak die pressure.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1997년도 추계학술대회논문집
• /
• pp.153-157
• /
• 1997

The investigation deals with the process design in cold rolling mill of electric plant. In this study, multiple objective optimization is conducted by a genetic algorithm, where the fitness values are evaluated on the basis of one - dimensional model of flat rolling. The approach is applied to the determination of the process conditions which are optimal with regard to minimization of roll power and maximization of productivity.

• 생물정신의학
• /
• 제26권2호
• /
• pp.33-38
• /
• 2019

Psychiatric disorder as dysfunctional behavioural syndrome is a paradoxical phenomenon that is difficult to explain evolutionarily because moderate prevalence rate, high heritability and relatively low fitness are shown. Several evolutionary genetic models have been proposed to address this paradox. In this paper, I explain each model by dividing it into selective neutrality, mutation-selection balance, and balancing selection hypothesis, and discuss the advantages and disadvantages of them. In addition, the feasibility of niche specialization and frequency dependent selection as the plausible explanation about the central paradox is briefly discussed.

• 디지털융복합연구
• /
• 제6권2호
• /
• pp.41-53
• /
• 2008

This research is to redefine the concept of Good School and to validate an effective Good Secondary School Scale in Kyung-gi Province and Seoul. As statistical methods, SPSS 13.0 and AMOS 5.0 were used. Item Analysis and Exploratory Factor Analysis(EFA) were conducted to test the reliability of items and the factor structure. And Confirmatory Factor Analysis(CFA) was conducted to test the validity and fitness of the Good School Scale. The outcomes are as follows: First, six factors(school environment, curriculum, teacher, school-based management system, director) will increase the good schooling effectiveness. Second, As a result of Confirmatory Factor Analysis(CFA), the goodness of fit indices(GFI AGFI, CFI, RMSEA) demonstrate statistically significance and fitness of the model. The final Good School Scale supports 6 Good School Factors obtained in main test. Therefore, we can say that this scale can be used as a valid instrument to measure a real Good Schooling Effectiveness at the secondary school situation in Korea.

• 한국인공지능학회지
• /
• 제5권1호
• /
• pp.18-28
• /
• 2017

Recently, there are many companies that use systems based on artificial intelligence. The accuracy of artificial intelligence depends on the amount of learning data and the appropriate algorithm. However, it is not easy to obtain learning data with a large number of entity. Less data set have large generalization errors due to overfitting. In order to minimize this generalization error, this study proposed DGA which can expect relatively high accuracy even though data with a less data set is applied to machine learning based genetic algorithm to deep learning based dropout. The idea of this paper is to determine the active state of the nodes. Using Gradient about loss function, A new fitness function is defined. Proposed Algorithm DGA is supplementing stochastic inconsistency about Dropout. Also DGA solved problem by the complexity of the fitness function and expression range of the model about Genetic Algorithm As a result of experiments using MNIST data proposed algorithm accuracy is 75.3%. Using only Dropout algorithm accuracy is 41.4%. It is shown that DGA is better than using only dropout.

• 융합정보논문지
• /
• 제8권1호
• /
• pp.95-103
• /
• 2018

본 논문에서는 보행속도, 주야간 인구변화, 보행경로에 따른 서비스 권역 등을 고려하여 대피시설 위치에 대한 적절성을 평가하고자 하였다. 보행속도는 선행연구사례를 통해서 1.6 m/s, 2.22 m/s를 정의하였고, 주야간 인구변화는 대시메트릭 매핑기법을 이용하여 인구를 추정하였다. 보행속도와 보행경도에 따른 대피시설 서비스 권역은 입지할당모형의 네트워크 분석을 실시하였다. 그 결과 시공간 유동인구 변화와 보행속도에 따라 일부 대피시설의 경우 수용능력에 한계가 있음을 알 수 있었고 이에 대한 대피시설 추가지정이 요구된다.

• International Journal of Advanced Culture Technology
• /
• 제5권2호
• /
• pp.74-81
• /
• 2017

Recently, there are many companies that use systems based on artificial intelligence. The accuracy of artificial intelligence depends on the amount of learning data and the appropriate algorithm. However, it is not easy to obtain learning data with a large number of entity. Less data set have large generalization errors due to overfitting. In order to minimize this generalization error, this study proposed DGA(Dropout Genetic Algorithm) which can expect relatively high accuracy even though data with a less data set is applied to machine learning based genetic algorithm to deep learning based dropout. The idea of this paper is to determine the active state of the nodes. Using Gradient about loss function, A new fitness function is defined. Proposed Algorithm DGA is supplementing stochastic inconsistency about Dropout. Also DGA solved problem by the complexity of the fitness function and expression range of the model about Genetic Algorithm As a result of experiments using MNIST data proposed algorithm accuracy is 75.3%. Using only Dropout algorithm accuracy is 41.4%. It is shown that DGA is better than using only dropout.

• 한국안전학회지
• /
• 제17권4호
• /
• pp.5-10
• /
• 2002

For fitness-for-service evaluation of high temperature plant components with defects, crack growth life must be assessed properly as indicated in the recent draft of API 579 code. Type 308 stainless steel has been widely used as a field weld material in the petrochemical industry. In this study, creep crack data of type 308 stainless steel are collected and re-analyzed using $C_t$ as a characterizing fracture parameter. A unique da/dt versus $C_t$ relationship was obtained despite of difference of creep deformation constant of the reviewed materials and specimen geometry of the tested specimens. The obtained results can be employed for crack growth life assessment and fitness-for-service evaluation for the cracks in high temperature components. It is also argued that since the effect of creep properties and other material variability on the creep crack growth behavior would be minor the obtained model may be applied for most of the 308 stainless steels.