Tree Pattern Aggregation for Scalable XML Data Dissemination

In this paper, we study the novel problem of subscription aggregation, where the subscriptions are specified with tree patterns which are a uesful subclass of XPath expressions. The challenge is to aggregate an input set of tree patterns into a smaller set of generalized tree patterns within a given space constraint such that the "loss in precision" due to aggregation is minimized. To resolve this problem, we propose a novel and efficient approach based on the Minimum Description Length principle.