Optimization of the Answering Method with Probability Conversion

The growth of the Internet has facilitated data collection on a large scale. User opinions and behavior are private information and need to be collected using a method that protects their confidentiality. In this paper, we develop a method of collecting information that employs probability conversion. We make clear the relations among the answer functions' variance, the number of respondents, confidence and accuracy. We also describe the answer functions that maximize anonymity.