部分木制約付き頻出飽和木パターン発見

抄録

Two critical bottle necks in mining frequent tree patterns from tree databases are the exponential number of mined patterns and the lack of user focus on the mining process. We propose, in this paper, an algorithm that solves the problems for unordered attribute trees by mining only the compact representation of tree patterns, i.e. closed tree patterns, and allows users to mine only trees of their interest by specifying subtree constraints. The experimental results show the efficiency of our algorithm.