A Summary Structure of Data Cube Preserving Semantics

摘要

The semantic relations among cells in data cube are more important for efficient query and OLAP. Normally the size of a data cube is very huge and relations among cells are very complicated so the semantic data cube is difficult to be realized. Based on quotient cube, Semantic Data Cube (SDC) structure is put forward in this paper. In SDC the lattice of cells is expressed as tree-hierarchy structure and each cell in lattice is replaced with its upper bound. The SDC depicts the lattice of cells concisely and preserves all the semantic relations among cells. Applying semantics to query answering and maintaining incrementally in SDC, the time of response and the cost of updating can be reduced greatly. Algorithms of constructing SDC, answering a query and maintaining incrementally in SDC are given. The experimental results show that the SDC is effective.