• 디지털융복합연구
• /
• 제10권6호
• /
• pp.219-224
• /
• 2012

통신 모바일 단말기에서 어휘 인식 시스템은 부정확한 어휘로부터 음소 특징을 추출하기 때문에 음소를 인식하지 못하거나 유사한 음소 오인식 오류로 인한 낮은 인식률의 문제점을 가진다. 이러한 문제를 해결하기 위해서, 본 논문에서는 입력 음소는 음소 유사율 처리를 통해 음소 사이의 거리를 측정하여 수치로 나타내고, 신뢰도 측정을 통하여 인식되어진 결과를 확인하는 시스템을 제안하였다. 이로 인해 부정확한 어휘 제공으로 인한 오인식 오류를 최소화하였으며 음소 유사율과 신뢰도를 이용하여 오류 보정율을 구하였다. 기존 방법인 에러 패턴 학습을 이용한 시스템과 의미기반을 이용한 시스템의 성능 평가 결과 2.7%의 인식 향상율을 보였다.

• 대한안전경영과학회지
• /
• 제8권5호
• /
• pp.253-264
• /
• 2006

This paper is to derive three reliability and eight maintainability measures after organizing total time. I propose inherent availability,.achieved availability and operational availability with the new and useful reliability and maintainability measures above. Especially, inherent availability measure extends the repairable system reliability models.

• 한국지능시스템학회논문지
• /
• 제11권4호
• /
• pp.293-301
• /
• 2001

본 논문은 신경회로망과 같은 경험적 모델에서 출력별로 신뢰 구간을 추정하는 세 가지 대표적인 방법을 검토하고, 검토한 방법을 뉴로-퍼지 모델에 적용하여 장단점을 비교 분석한다. 본 논문에서 고려한 출력별 신뢰 구간 계산 방법은 cross-validation을 이용한 stacked generalization, 회귀 모델에서 유도된 predictive error bar, 지역 표현하는 신경회로망의 특성에 기반한 local reliability measure이다. 간단한 함수 근사화 문제와 혼돈 시계열 예측 문제를 이용하여 모의 실험을 수행하고, 세 가지 신뢰도 추정 방법의 성능을 정량적, 정성적으로 비교 분석한다. 분석 결과를 기초로 각 방법의 장단점 및 특성을 고찰하고, 모델링 문제에서 모델의 출력별 신뢰 구간 계산 방법의 실제 적용 가능성을 탐색한다.

• 한국전문물리치료학회지
• /
• 제26권1호
• /
• pp.60-66
• /
• 2019

Background: Knee osteoarthritis (OA) is a single most arthritic disease. Knee joint space width (JSW) is commonly used for grading severity of knee OA. However, previous studies did not established criterion validity and test-retest reliability of ultrasound (US) image for measuring JSW. Objects: The aim of this study was to establish criterion validity and test-retest reliability of US measurement of medial and lateral knee JSW. Methods: Twenty-nine subjects with knee OA were participated. The US and X-ray were used to measure knee JSW. One sample Kolmogorov-Smirnov test was used to confirm the data normal distribution. Pearson correlation coefficient and ICC were used to calculated and establish criterion validity and test-retest reliability, respectively. Results: US measurement of medial and lateral knee JSW was highly correlated with radiographic imaging measure (r=.714 and .704, respectively). Test-retest reliabilities of medial and lateral knee JSW were excellent correlated (ICC=.959 for medial side and .988 for lateral side, respectively). Conclusion: US may be valid tool to measure knee JSW.

• 한국병원경영학회지
• /
• 제11권3호
• /
• pp.42-55
• /
• 2006

The present study is designed to develop surveying tools to measure stressors of and social supports for Radiological Technologists which may play a role to alleviate their stresses caused from their job environments. For this purpose, an analysis of reliability by factors of personal stressors was conducted on 890 radiologists who are working at 44 general hospitals in 16 cities and provinces across the country. The results of the analysis may be summed up as follows: Six factors in 2 concerned fields turned out to be major tools for use as items of questionnaires to measure social supports for and organizational culture of Radiological Technologists. The factors determined to be applicable for survey with reliability coefficient of more than 0.7 measured by factors for internal coherence included social supports on the job(0.90), off-the-job social supports(0.81), job satisfaction(0.80), devotion for job by length of service(0.81) and emotional devotion for job(0.80), all deemed to be appropriate as survey items. In the verifying process of the results, it was found that normative devotion for job(0.61) showed lower reliability as its statistical coefficient was analyzed to be less than 0.7. It seemed, therefore, necessary to reconstruct items or re-select such ones only that can maintain internal coherence in order to analyze appropriateness of related measuring tools, or to develop new measuring tools. Although the outcomes of this study can be used to verify reliability and appropriateness of concerned concepts to assess stresses caused during the job performance of Radiological Technologists, it may be improper to adopt them to measure all types of social supports and stresses that occur generally under the job environment. It is desired, however, to compare and analyze the findings of the present study with others and to utilize them as basic data for further development of tools that can maintain internal coherence in terms of reliability and appropriateness in spite of changing social environments.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• 제33권3호
• /
• pp.67-72
• /
• 2022

Objectives: This study investigated the reliability and validity of the Korean version of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) Level 2 Cross-Cutting Symptom Measure-Patient-Reported Outcomes Measurement Information System (PROMIS)-Depression and the Irritability for parents of children aged 6-17 years. Methods: Participants were 190 children diagnosed with depressive disorder (n=14), anxiety disorder (n=21), attention-deficit/hyperactivity disorder (ADHD; n=111), ADHD with anxious depression (n=13), and tic disorder with somatic symptoms (n=31). Patients were 8-15 years of age. The participants' mothers completed the Korean versions of the DSM-5 Level 2 Cross-Cutting Symptom Measure-PROMIS Depression and Irritability (Affective Reactivity Index, ARI), and the Korean Child Behavior Checklist (K-CBCL). Using these data, we calculated the reliability coefficient and examined the concurrent and discriminant validity of the PROMIS Depression and the Irritability (ARI) scales for assessing depression and irritability in children. Results: The reliability coefficient of the PROMIS Depression scale (Cronbach's α) was 0.93. The correlation coefficient with the K-CBCL DSM emotional problem score was 0.71. The PROMIS Depression scale significantly discriminated children with depressive disorders from those with other conditions. The reliability coefficient of the Irritability (ARI) scale was 0.91, suggesting its high reliability. Conclusion: Our results suggest that the Korean version of the DSM-5 Level 2 Cross-Cutting Symptom Measure for Depression and Irritability Scales for parents of children aged 6-17 years is reliable and valid and may be an efficient alternative to the K-CBCL.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제5권2호
• /
• pp.203-207
• /
• 2005

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제5권1호
• /
• pp.143-148
• /
• 2005

• Genomics & Informatics
• /
• 제9권1호
• /
• pp.28-36
• /
• 2011

Most statistical methods for finding interesting genes are focusing on the summary values with large fold-changes or large variations. Very few methods consider the probe level data. We developed a new measure to detect reliability that incorporates the probe level data. This reliability measure is useful for exploring the microarray data without ignoring the probe level data. It is easy to calculate, and it can be used for all the other statistical methods as a good guideline to find real differentially expressed genes. Instead of filtering out genes before the analysis, we use whole genes in the analysis and make decisions with new reliability measures.

• Asia pacific journal of information systems
• /
• 제11권4호
• /
• pp.81-98
• /
• 2001

This study investigates the reliability and validity of research instruments used in Korean MIS research. Examination of reliability and validity is a necessary process for establishing proper arguments based on research results. We examined 172 MIS research articles published in four major Korean journals between 1990 and 1999. Among them, 79 papers which used multivariate research methods were analyzed in this study. The relationships between reliability measures and several research design characteristics were examined. We found that scale format and the number of items used to measure a construct were significantly correlated with reliability measure. About 27 percent of the articles we examined did not perform validity analysis. Based on the findings in our study, we proposed a checklist for presenting data validation analysis.