Using Tree Automata for XML Mining and Web Mining with Constraints

Most work on pattern mining focus on simple data structures like itemsets or sequences of itemsets. However, a lot of recent applications dealing with complex data like chemical compounds, protein structure, XML and Web Log databases and social network, require much more sophisticated data structures (trees or graphs) for their specification. Here, interesting patterns involve not only frequent object values (labels) appearing in the graphs (or trees) but also frequent specific topologies found in these structures. In a recent work, we have introduced the method CobMiner for tree pattern mining with constraints. CobMiner uses tree automata as a mechanism to specify user constraints over tree patterns. In this paper, we focus on applications of this method in two different contexts: XML Mining and Web Usage Mining. An extensive set of experiments executed over real data in these two fields allow us to conclude that CoBMiner is an efficient method for mining datasets whose elements are specified as trees.