Schema Discovery of Semi-structured Hierarchical Data Based on OEM Model and Hierarchical Transactional Database

Semi-structured data is different from the traditional data model. It has data prior to schema, and semi-structured data model is used to describe the data structural information of data rather than the mandatory constraint. Therefore, discovery of semi-structured data has become the first step of knowledge discovery. In this paper, the concept of hierarchical data is adopted and a counting principle of "cumulative transformation" and hierarchical transactional database is described. Accordingly, a SHDP-mine mining algorithm based on SHDP-tree and a basic schema for data mining of semi-structured hierarchical data is presented. Finally, its effectiveness and efficiency is validated on a theoretical level and by experimental analysis.