We propose a relational database model towards the integration of data mining into relational database systems, based on the so called virtual mining views. We show that several types of patterns and models over the data, such as itemsets, association rules, decision trees and clusterings, can be represented and queried using a unifying framework. We describe an algorithm to push constraints from SQL queries into the specific mining algorithms. Several examples of possible queries on these mining views, using the standard SQL query language, show the usefulness and elegance of this approach.
|Title of host publication||Proceedings of the 1st International Workshop on Constraint-Based Mining and Learning (CMILE'07), 21September 2007, Warsaw, Poland|
|Editors||S. Nijssen, L. De Raedt|
|Place of Publication||Leuven|
|Publisher||Katholieke Universiteit Leuven|
|Publication status||Published - 2007|