Home // DBKDA 2013, The Fifth International Conference on Advances in Databases, Knowledge, and Data Applications // View article

FIMIOQR: Frequent Itemsets Mining for Interactive OLAP Query Recommendation

Authors:
Rym Khemiri
Fadila Bentayeb

Keywords: Interactive Recommendation; Data warehouse; Decision Query; Measure; Dimension attribute; Frequent Itemsets Mining; OLAP; Data warehouse

Abstract:
We propose in this paper an interactive query recommendation system, namely FIMIOQR. It is designed to help OLAP (On-line Analytical Processing) users in their decision query writing task based on both a set of selected measures and decision queries log file. Our FIMIOQR system is designed to discover associations from decision queries log file. For this end, we use association rules method to extract frequent itemsets from dimensions attributes according to user selected set of measures. This allows end users in OLAP systems to write relevant queries guided by an interactive recommending system and helps them to meet their analysis objectives. In addition, we propose a tool for the automatic implementation of FIMIOQR which provides a visual interface to OLAP users which helps them to write their queries step by step in an interactive way. We also carried out some experimental tests to evaluate our system. The experimental evaluation proves our FIMIOQR framework is efficient in term of recommendation quality.

Pages: 9 to 14

Copyright: Copyright (c) IARIA, 2013

Publication date: January 27, 2013

Published in: conference

ISSN: 2308-4332

ISBN: 978-1-61208-247-9

Location: Seville, Spain

Dates: from January 27, 2013 to February 1, 2013