• Journal of the Korean Data and Information Science Society
• /
• 제24권5호
• /
• pp.1029-1041
• /
• 2013

최근 빅데이터의 등장과 정보 공개에 대한 급격한 수요 증가에 따라 자료를 일반에게 공개할 때 개인 정보를 보호해야 하는 필요성이 어느 때보다 절실하다. 본 논문에서는 마이크로 자료와 통계분석 서버를 중심으로 현재까지 제시된 개인정보 노출제한를 위한 통계적 방법, 정보 노출의 개념, 노출 위험을 측정하는 기준들을 개괄적으로 소개한다.

• 응용통계연구
• /
• 제22권1호
• /
• pp.141-151
• /
• 2009

제공되는 마이크로자료가 어떤 형태인지에 따라 응답자의 개인정보를 보호하는 방법도 다르게 적용된다. 본 연구에서는 연속형자료의 비밀보호에 효과적인 잡음(noise)을 이용하는 방법을 소개하고, 통계청에서 실시한 2005년 가계조사 자료에 이 방법을 적용하여 응답자의 정보노출이 제한된 마이크로자료를 작성하는 과정을 설명한다. 잡음의 생성을 위해 삼각분포와 절단된 삼각분포, 사다리꼴분포 그리고 이중삼각분포를 이용하고 소지역 추정에 필요한 공식도 유도한다. 아울러 각 분포별로 얻어진 잡음을 이용하여 가계조사 자료를 변환하여 비교 분석한 결과도 보여준다.

• 대한기관윤리심의기구협의회지
• /
• 제2권1호
• /
• pp.23-31
• /
• 2020

Purpose: To obtain opinions from Korean Institutional Review Board (IRB) members' self-evaluation on ability to conduct fairness review of clinical trial protocol with presence of conflict of interest and from investigators and IRB members on financial conflict of interest through surveying. Methods: IRB members and researchers in 9 different hospitals were asked to answer survey questions via email. Results: Responders were 115 personnel (IRB Chair/vice 18, medical member 30, non-medical member 28, and researcher 39) from 9 centers. Compared to IRB medical members, IRB chair/vice respondents scored higher with statistically significance on 10 point scale (8.44±1.381 vs. 7.30±1.685, p=0.005) when asked to self-evaluate fairness reviewing a protocol proposed by an investigator from the same department and a protocol from the company that supports the scientific committee of responders. When reviewing a protocol proposed by a hospital director, non-medical members scored statistically significantly higher than medical-members (7.47±1.76 vs. 8.07±2.70, p=0.034). When asked about the limitation of labor fee for principal investigator on phase 3 Human clinical trials of the Investigational new drug, while the responses range was wide, 60% answered that labor cost of principal investigator should be less than 30% of total budget for clinical trials with a budget of 100 million won. 51.3% answered that there is no need to disclose the labor cost of the principal investigator in the consent form. Since every investigator can be influenced unconsciously by conflict of interest, the answer that 'responder agrees that there is need for management' was the most chosen answer (IRB member 61.8%, investigator 64.1%, multiple answers allowed). Conclusion: Considering scores on questions of fairness by IRB members were between 7.23-8.56 on scale of 0 to 10 point when IRB members were asked about reviewing a clinical trial protocol, it cannot be said with absolute certainty that there is no issue regarding fairness in the review process. Therefore, there should be more ways to safeguard fairness for these issues. There is a need that the disclosure amount of honorarium from sponsor should be lower than 100 million Korean won. Considering the results of the survey in which respondents expressed their thoughts, it is likely that more education on the concept of conflict of interest is needed.