A Data Cube Representation for Efficient Querying and Updating

Abstract : —This paper presents an approach to data cube representation that allows for efficient computation of query response and data cube update. The representation is based on (i) a recursive construction of the power set of the fact table dimension scheme and (ii) a prefix tree structure for the compact storage of cuboids. Experimental results on large datasets show that the approach is efficient for data cube query and incremental update. Keywords-data warehouse; data cube; data mining;