Providing probabilistically-bounded approximate answers to non-holistic aggregate range queries in OLAP

A novel framework for providing probabilistically-bounded approximate answers to non-holistic aggregate range queries in OLAP is presented in this paper. Such a framework allows us to efficiently support OLAP applications, as answering queries is the main bottleneck for this kind of applications. To this end, scalability of the techniques and accuracy of the answers are recognized as important limitations of state-of-the-art approximate query answering proposals in OLAP. Specifically, this paper is focused on the latter limitation, whereas it refers to results presented in [9] for the first one. The KSyn synopsis data structure, which implements the guidelines of the proposed framework and overcomes the recognized limitations, is also presented and discussed in detail, along with a query-conscious error metrics-based storage space allocation scheme. Finally, encouraging preliminary experimental results stating the goodness of our proposal are presented and discussed. Categories and Su...

• Type: Conference
• Year: 2005
• Where: DOLAP
• Authors: Alfredo Cuzzocrea
• Keywords: Approximate Query, DOLAP 2005, Information Management, Probabilistically-bounded Approximate Answers, Synopsis Data Structures