Skip to main content
k-anonymization is a method of processing data to make it difficult to identify individuals. The method has become increasingly important as a means to protect privacy in accordance with the expanding uses of data.
By converting the object data so that there are at least k records of data with identical attributes (meaning that the k-anonymity requirement is satisfied), the method reduces the probability of any individual being identified to 1/k or less. A variety of k-anonymization algorithms are currently available, but they have issues such as that a large volume of information is lost through anonymization and it takes time to process data. Improvements for these issues are ongoing.
As applications of k-anonymization, proposals have been made on methods to make identification of individuals more difficult, including a method to process data so that there is at least l type of attribute (meaning that the l-diversity requirement is satisfied) and a method to process data so that the deviation in data distribution is reduced (meaning that the t-closeness requirement is satisfied).