• 제목/요약/키워드: Data Quality Problem Analyze

검색결과 139건 처리시간 0.023초

품질문제 요인분석을 통한 데이터 품질 개선방안 연구:국가R&D정보연계 방식 중심 (Research of Quality Improvement by Factors Analysis Data Quality Problem:Focus on National R&D Information Linking Structure)

  • 손강렬;임종태
    • 한국콘텐츠학회논문지
    • /
    • 제9권1호
    • /
    • pp.1-14
    • /
    • 2009
  • 현재 국내의 정부 R&D 사업은 300여개에 이르고 있고, 이를 사업의 특성별로 16개 국가R&D 관련 부처 청의 15개 대표연구관리전문기관에서 각각 관리하고 있다. 이로 인하여 발생하는 국가 R&D에 대한 중복 투자와 체계적인 R&D연구과제 및 성과관리의 미흡으로 R&D 투자의 효율성에 대한 문제가 계속해서 제기되고 있다. 그러한 가운데 이러한 문제를 해결하기 위하여 교육과학기술부는 국가연구개발의 기획에서 성과활용에 이르기까지 연구개발의 효율화를 지원할 수 있는 국가 R&D 포털시스템으로써 국가과학기술종합정보시스템(NTIS)을 구축하고 있다. 본 논문에서는 NTIS시스템의 인력/과제/성과 정보의 통합 DB 구축 및 연계방식과 이를 통해 수집된 데이터의 품질관리를 위한 데이터 정제 프로세스를 고찰해 본다. 그 과정에서 발생할 수 있는 데이터 품질문제의 요인을 분석하여 NTIS의 데이터 품질향상을 위한 개선방안을 제시한다.

와이블 수명자료들에 대한 베이지안 가설검정 (Bayesian Hypotheses Testing for the Weibull Lifetime Data)

  • 강상길;김달호;조장식
    • 품질경영학회지
    • /
    • 제28권3호
    • /
    • pp.1-10
    • /
    • 2000
  • In this paper, we address the Bayesian hypotheses testing for the comparison of Weibull distributions. In Bayesian testing problem, conventional Bayes factors can not typically accommodate the use of noninformative priors which are Improper and are defined only up to arbitrary constants. To overcome such problem, we use the recently proposed hypotheses testing criterion called the intrinsic Bayes factor. We derive the arithmetic and median intrinsic Bayes factors for the comparison of Weibull lifetime model and we use these results to analyze real data sets.

  • PDF

Hypotheses Testing for the Shape Parameter of the Weibull Lifetime Data

  • Kang, Sang-Gil;Kim, Dal-Ho;Cho, Jang-Sik
    • 품질경영학회지
    • /
    • 제27권4호
    • /
    • pp.153-166
    • /
    • 1999
  • In this paper, we address the Bayesian hypotheses testing for the shape parameter of weibull model. In Bayesian testing problem, conventional Bayes factors can not typically accommodate the use of noninformative priors which are improper and are defined only up to arbitrary constants. To overcome such problem, we use the recently proposed hypotheses testing criterion called the intrinsic Bayes factor. We derive the arithmetic and median intrinsic Bayes factors and use these results to analyze real data sets.

  • PDF

NASA MDP 데이터 집합의 결함도 모호성 분석 (Ambiguity Analysis of Defectiveness in NASA MDP Data Sets)

  • 홍의석
    • 한국IT서비스학회지
    • /
    • 제12권2호
    • /
    • pp.361-371
    • /
    • 2013
  • Public domain defect data sets, such as NASA data sets which are available from the NASA MDP and PROMISE repositories, make it possible to compare the results of different defect prediction models by using the same data sets. This means that repeatable and general prediction models can be built. However, some recent studies have raised questions about the quality of two versions of NASA data set, and made new cleaned data sets by applying their data cleaning processes. We find that there are two ways in the NASA MDP versions to determine the defectiveness of a module, 0 or 1, and the two results are different in some cases. This serious problem, to our knowledge, has not been addressed in previous studies. To handle this ambiguity problem, we define two kinds of module defectiveness and two conditions that can be used to determine the ambiguous cases. We meticulously analyze 5 projects among the 13 NASA projects by using our ambiguity analysis method. The results show that JM1 and PC4 are the best projects with few ambiguous cases.

대학수학능력시험 점수산정시스템의 품질 제고를 위한 연구 (A Study on the Quality Improvement of College Scholastic Ability Test Scoring System)

  • 박영선
    • 품질경영학회지
    • /
    • 제50권2호
    • /
    • pp.199-220
    • /
    • 2022
  • Purpose: The purpose of this study is to analyze the score data released by the Korea Institute of Curriculum and Evaluation to find out the problems with the current scoring system provided by the College Scholastic Ability Test and to suggest improvement measures to solve these problems. Methods: We calculated the descriptive statistics of the standard scores using the frequency distribution table of the standard scores and identified the characteristics of the standard scores by expressing the distribution as a graph. Also, we developed an index to evaluate whether each stanine level was stably assigned and calculated the indexes for each area/subject by using the data on the number of examinees for each level. Results: We found that the relationship of conversion from raw scores to integerized standard scores is different depending on the size of the standard deviation of the raw scores, and identified the problem that the raw score information is not fairly reflected in the calculation of the percentile and level as the two raw scores are converted to one standard score. This problem can be solved by calculating the standard score to a decimal point. Conclusion: In this study, as a way to improve the quality of the scores of the current CSAT, the standard score and percentile decimal notation, the specific regulations of the standard score and stanine level calculation method, and the expansion of the open range of the scores were suggested.

수질 모니터링 데이터 기반의 수질센서 자가진단 알고리즘 (Self-diagnosis Algorithm for Water Quality Sensors Based on Water Quality Monitoring Data)

  • 김홍중;김종민;강태형;류갑상
    • 사물인터넷융복합논문지
    • /
    • 제9권1호
    • /
    • pp.41-47
    • /
    • 2023
  • 오늘날, 세계 인구성장률의 증가로 국제사회는 심각하게 식량문제 해결을 논의하고 있다. 식량문제 해결을 위한 대안으로는 양식산업이 대두되고 있다. 최근 양식산업의 혁신성장을 위해 4차 산업기술을 융합한 스마트 양식장이 보급되고 있으며, 전주기적 디지털화가 추진되고 있다. 양식산업에서 중요한 수질센서는 전기화학방식의 휴대용 센서를 사용하고 있으며, 이를 이용하여 개별적, 간헐적으로 수질을 체크하고 있어서 양식장 수질을 실시간 분석하고 관리하기가 불가능하다. 최근 광학 기반의 모니터링이 가능한 수질센서들이 개발되어 현장에 적용되고 있다. 그러나 수질센서의 상태정보를 알 수 없기 때문에 모니터링 데이터의 신뢰성을 보장할 수 없는 상황이다. 따라서, 본 논문에서는 데이터의 신뢰성을 확보할 수 있도록, 수질센서가 수집하는 모니터링 데이터를 기반으로 고장, 기준일탈, 유지보수, 점검 등의 수질센서 자가진단 상태를 파악할 수 있는 알고리즘을 제안한다.

Perceived Distribution Quality Awareness, Organizational Culture, TQM on Quality Output

  • ISNAINI, Dewi Budhiartini Juli;DANILWAN, Yuris;MANSUR, Daduk Merdika;ILYAS, Gunawan Bata;MURTINI, Sri;TAUFAN, Muhammad Ybnu
    • 유통과학연구
    • /
    • 제19권12호
    • /
    • pp.1-14
    • /
    • 2021
  • Purpose: For the last few decades, TQM has become a hot topic in the inner-disciplinary field in the production management line. Still, unfortunately, the study of TQM and Quality Output management is partially only attached to the tangible side in the production management line. Whereas theoretically, the implications of TQM require incremental improvement in all management lines (e.g., HRM, Marketing, Operations, and Distribution Management). Therefore, starting from the main problem, this study aims to analyze the effect of total quality management, Organizational Culture, Perceived Distribution of Quality Awareness, and quality output through a more in-depth analysis. Research design, data and methodology: We conducted a survey of 170 respondents from managers, staff, and employees from 48 companies in Indonesia. We used a quantitative approach with the SEM method to answer this study's problem formulation and hypotheses. Results: The results of our research stated that based on the demonstration of statistical test results, all hypotheses were positive and significant, both direct and indirect relationship demonstrations. Conclusions: Universally, the findings in this study illustrate that the supporting factor for creating value-added in TQM and Quality output lies in the optimal and positive organizational culture and Perceived Distribution Quality Awareness factors in the organization.

정보화사회에서의 소비자 문제에 관한 연구 -인터넷 상거래를 중심으로- (Consumer Problems in the Information Society -Focus on Internet Commerce-)

  • 이승신
    • 대한가정학회지
    • /
    • 제39권5호
    • /
    • pp.15-26
    • /
    • 2001
  • The purpose of this study is to examine specific consumer problems in internet commerce and analyze them in the age of information. The way of sorting consumer problems in internet commerce from this study. For this empirical analysis, the data was collected 319 consumers by internet on-line survey from Feb. 13 to Feb. 20, 2000. The methods for study was frequency, percentage, mean, correlation, multiple regression using SAS PC program. The major findings of this study were as follows: 1. The level of consumer problem experiences in internet commerce was examined. The problems of privacy was the most problem and the problem of price was the second. Then delivery, advertisement, information was next, and the problem of quality, and A/S, change, cancel was the next consumer problem. 2. Among factors influence on experience consumer problem in internet commerce, the amount of usable money, information search were significant.

  • PDF

현장실사에 의한 의료 서비스품질 개선방안 모색 (Health-care Service Quality Improvement Using Walk-through Audit)

  • 류문찬;신지연
    • 품질경영학회지
    • /
    • 제41권4호
    • /
    • pp.527-539
    • /
    • 2013
  • Purpose: Methods for increasing health-care service quality are considered for a women-centered hospital located on a local city. A walk-through audit is applied to uncover areas for improvement. Methods: A survey questionnaire with 25 questions is constructed based on a service blueprint for a walk-through audit, and a survey is conducted both to patients and service providers to assess a health-care service quality. Frequency analysis, statistical tests and customer-provider analysis are used to analyze surveyed data. Results: According to customer-provider analysis, 6 attributes belong to 'problem unawareness' zone in which they are rated high by service providers but low by patients, and another 6 attributes belong to 'problem awareness' zone in which they are rated low by both groups. These attributes are considered to be improved with priority. Conclusion: Both patient group and service provider group have lots of different perceptions on most attributes that are examined, and the hospital to be studied is, in general, competitive in technical quality and less competitive in functional quality.

문제중심학습을 적용한 노인보건교육 프로그램의 효과 (Effects of a Problem-Based Learning Program on Health Education for Elders)

  • 손영주;최은영;송영아
    • 대한간호학회지
    • /
    • 제39권2호
    • /
    • pp.207-216
    • /
    • 2009
  • Purpose: The purposes of this study was to analyze the effects of a health education program using problem-based learning on health related knowledge, behavior, and quality of life in elderly people. Methods: The participants included 44 elders, of whom 23 took the health education program and 21 did not. All participants were over 60 yr of age and were selected from residents of nursing homes or participants in activities of social welfare facilities in Jeju Province. Elders in both groups completed pre- and post-tests. Elders in the education group participated in 5 weekly sessions, 100-120 min/session of problem-based learning on health education. Data were analyzed using SPSS WIN 12.0. Results: Scores for health knowledge, health behavior, and quality of life for the education group were significantly higher than those of the control group. Conclusion: A problem-based learning health education program can be recommended as a method to promote the health of the elders. Indirectly, the results seem to indicate that proper assessment and support should be provided simultaneously in the management of elders' health. Finally, future study is needed to examine whether problem-based learning is more helpful compared to traditional education.