Getting right answers from incomplete multidimensional databases

Dealing with large volumes of data, OLAP data cubes aggregated values are often spoiled by errors due to missing values in detailed data. This paper suggests to adjust aggregate answers, noticing that non-missing values constitute a biased sample of the true result of the query. Using basic random sampling theory, we show that two different problems can be solved nicely: (1) the case of missing tuples in the database, (2) the case of missing values appearing in the attributes used to build the data cube dimensions. Integration of these concepts within the OLAP data cube model is solved, by adjusting the data cube measures with a well-chosen weighting system. An algorithm (the ROWN method) minimizes the number of necessary weighting systems. A proof of concept implementation on the ORACLE EXPRESS system is briefly described at the end of the paper. RESUME. Dans le contexte des OLAP, les valeurs manquantes au sein des donn

• Type: Journal
• Year: 2002
• Where: ISI
• Authors: Sabine Goutier, Georges Hébrail, Véronique Stéphan
• Keywords: Cube, ISI 2002, Missing Values, OLAP Data, Security Privacy