SQL extension for exploring multiple tables

The standard SQL assumes that the users are aware of all tables and their schemas to write queries. This assumption may be valid when the users deal with a relatively small number of tables, but writing a SQL query on a large number of tables is often challenging; (1) the users do not know what tables are relevant to their query, (2) it is too cumbersome to explicitly list tens of (or even hundreds of) relevant tables in the FROM clause and (3) the schemas of those tables are not identical. In this paper, we propose an intuitive yet powerful extension to SQL that helps users explore and aggregate information spread over a large number of tables. With our extension, users can declaratively specify the tables of interest using the concept of tablesets, as they can declaratively specify the rows of interest by boolean conditions with the standard SQL. Seven primitive operators on tablesets are investigated for creating, manipulating, and aggregating data for tablesets. Our user study sho...

• Type: Conference
• Year: 2008
• Where: CIKM
• Authors: Sung Jin Kim, Junghoo John Cho
• Keywords: CIKM 2008, Information Management, Large Number, SQL Query, Standard Sql