Query Recommendations for OLAP Discovery-Driven Analysis
Giacometti, Arnaud; Marcel, Patrick; Negre, Elsa; Soulet, Arnaud (2009), Query Recommendations for OLAP Discovery-Driven Analysis, in Song, Il-Yeol; Zimanyi, Esteban, DOLAP 2009, ACM 12th International Workshop on Data Warehousing and OLAP, Hong Kong, China, November 6, 2009, Proceedings, ACM, p. 81-88
TypeCommunication / Conférence
Conference title12th International Workshop on Data Warehousing and OLAP DOLAP 2009 (in conjunction with ACM CIKM 2009)
Conference cityHong Kong
Book titleDOLAP 2009, ACM 12th International Workshop on Data Warehousing and OLAP, Hong Kong, China, November 6, 2009, Proceedings
Book authorSong, Il-Yeol; Zimanyi, Esteban
MetadataShow full item record
Abstract (EN)Recommending database queries is an emerging and promising field of research and is of particular interest in the domain of OLAP systems, where the user is left with the tedious process of navigating large datacubes. In this paper, the authors present a framework for a recommender system for OLAP users that leverages former users’ investigations to enhance discovery-driven analysis. This framework recommends the discoveries detected in former sessions that investigated the same unexpected data as the current session. This task is accomplished by (1) analysing the query log to discover pairs of cells at various levels of detail for which the measure values differ significantly, and (2) analysing a current query to detect if a particular pair of cells for which the measure values differ significantly can be related to what is discovered in the log. This framework is implemented in a system that uses the open source Mondrian server and recommends MDX queries. Preliminary experiments were conducted to assess the quality of the recommendations in terms of precision and recall, as well as the efficiency of their on-line computation.
Subjects / KeywordsMDX queries; collaborative filtering; query recommendation; discovery-driven analysis; OLAP
Showing items related by title and author.