Efficient Algorithms for Discovering Frequent and Maximal Substructures from Large Semistructured Data

In this paper, we review recent advances in efficient algorithms for semi-structured data mining , that is, discovery of rules and patterns from structured data such as sets, sequences, trees, and graphs. After introducing basic definitions and problems, We present efficent algorithms for frequent and maximal pattern mining for classes of sets, sequences, and trees. In particular, we explain general techniques, called the rightmost expansion and PPC-extension, which are powerful tools for designing efficient algorithms. We also give examples of applications of semi-structured data mining to real world data.