China
Africa
Explore chapters and articles related to this topic
Collection of Data
Published in Shyama Prasad Mukherjee, A Guide to Research Methodology, 2019
In direct response surveys, the sample size required to estimate a population parameter with a specified error in estimation can be found from the condition Prob. [| t − Y | ≤ f Y] ≥ 1 − α where t is an unbiased estimator of the parameter Y, α is a fraction usually quite small and α is a pre-assigned small quantity., which gives n = (N− 1) / N CV2 / α f² for SRSWR and = N / [1 + N α f² (100 / CV²)] for SRSWOR where CV is the coefficient of variation of the estimator t (as given by Chaudhuri in 2010, 2014 and Chaudhuri and Dutta in 2018). The expression for sample size for general sampling strategies has also been indicated by Chaudhuri and Dutta (2018). A solution to the sample size problem in the case of a randomized response survey is yet to be found.
Event Log Privacy Based on Differential Petri Nets
Published in Applied Artificial Intelligence, 2023
Daoyu Kan, Xianwen Fang, Ziyou Gong
According to the concept of differential privacy, in order to achieve the protection of the private information of the case entity, i.e., the patient, in the event log, it is first necessary to randomize the important labels in this Petri net model to achieve anonymity. In this article, a randomized response algorithm is introduced to randomize the important labels. In life, we often encounter the situation that in a survey, when a question involves the privacy of the respondent, it may happen that the respondent does not want to give the true answer, and a solution is to let each respondent add noise to their respective answers. For example, suppose the interviewer asks a sensitive question about right and wrong, the respondent can flip a coin once and answer the true answer if the result of the coin flip is heads, or randomly answer yes or no if it is tails, which is the concept of randomized response. The process of obtaining the differential Petri net corresponding to a case in the event log using the randomized response algorithm is shown in Figure 4.
Deceptive Infusion of Data: A Novel Data Masking Paradigm for High-Valued Systems
Published in Nuclear Science and Engineering, 2022
Arvind Sundaram, Hany Abdel-Khalik, Ahmad Al Rashdan
Another well-known example of local differential privacy is that of a randomized response when dealing with surveys of illegal behavior. For example, consider the case of estimating the probability of cheating in university exams. A randomized response procedure would ask survey takers to toss a coin and answer the question honestly if heads and simply answer an arbitrary response otherwise. This protects the survey taker from releasing potentially incriminating information while providing sufficient information to predict the probability of cheating during an exam.