Data desensitization method for class information

Abstract

The invention discloses a data desensitization method for class information. The data desensitization method for the class information includes the following steps that 1) the class information is obtained, and ASCII values corresponding to control characters are obtained according to an ASCII comparison table; 2) decimal integers are randomly generated, the decimal integers are converted into numbers with the same scale with the ASCII values in the step 1), and then the obtained numbers and the ASCII values in the step 1) are summed; 3) a dictionary is established, wherein keys of the dictionary are characters between 0-9 and a-f, and values in the dictionary are 16 randomly-selected non-repetitive English capital letters; 4) every summation operation result serves as corresponding valuessearched by keys in the dictionary established in the step 3), then all the searched values are combined in sequence to form character strings, and finally, the character strings serve as data desensitization results for the class information. By means of the data desensitization method for the class information, the frequency and the percentage of the class information after desensitization canbe reserved, and data analysis values are not reduced.

Description

technical field [0001] The invention belongs to the technical field of privacy protection security, and relates to a data desensitization method for category information. Background technique [0002] With the elementization of data production, the continuous development of data science and data technology, and the in-depth mining and application of data value, a big data revolution is underway, and all walks of life are generating a huge number of data fragments every day. At present, a large amount of sensitive data such as customer category information has been accumulated in the business production system, such as whether a certain disease (Y / N), the degree of cure of a certain disease (H / M / L), etc. are very important Private information, once these private information is stolen and used by criminals, will bring economic and even reputation losses to individuals. Therefore, data owners must desensitize category information when using customer information. At present, t...

AI Technical Summary

Problems solved by technology

In the prior art, the desensitization method for category information is commonly used to be fixed replacement, that is, to replace all category information with the same fixed value. In this way, the frequency and percentage of category information are completely changed, so that the category information of the overall data set is lost. The value of data analysis

Images