Approximation Algorithms for Min-Max Generalization Problems

We provide improved approximation algorithms for the min-max generalization problems considered by Du, Eppstein, Goodrich, and Lueker [Du et al. 2009]. Generalization is widely used in privacy-preserving data mining and can also be viewed as a natural way of compressing a dataset. In min-max generalization problems, the input consists of data items with weights and a lower bound w_lb, and the goal is to partition individual items into groups of weight at least w_lb while minimizing the maximum weight of a group. The rules of legal partitioning are specific to a problem. Du et al. consider several problems in this vein: (1) partitioning a graph into connected subgraphs, (2) partitioning unstructured data into arbitrary classes, and (3) partitioning a two-dimensional array into contiguous rectangles (subarrays) that satisfy these weight requirements.

We significantly improve approximation ratios for all the problems considered by Du et al. and provide additional motivation for these problems. Moreover, for the first problem, whereas Du et al. give approximation algorithms for specific graph families, namely, 3-connected and 4-connected planar graphs, no approximation algorithm that works for all graphs was known prior to this work.