https://www.researchgate.net/publication/221000893_Capturing_data_usefulness_and_privacy_protection_in_K-anonymisation

Loukides and Shao [7] proposed another greedy algorithm for k-anonymization. Similar to the k-member algorithm, this algorithm builds one cluster at a time. But, unlike the k-member algorithm, this algorithm chooses the seed (i.e., the first selected record) of each cluster randomly. Also, when building a cluster, this algorithm keeps selecting and adding records to the cluster until the diversity (similar to information loss) of the cluster exceeds a user-defined threshold. Subsequently, if the number of records in this cluster is less than k, the entire cluster is deleted. With the help of the user-defined threshold, this algorithm is less sensitive to outliers. However, this algorithm also has two drawbacks. First, it is difficult to decide a proper value for the user-defined threshold. Second, this algorithm might delete many records, which in turn cause a significant information loss.