Conqueror is an algorithm for mining interesting association rules of frequent queries in arbitrary relational databases. It mines the simple, but appealing subclass of simple conjunctive queries. The Conqueror algorithm makes use of known functional dependencies in order to avoid the gener- ation of redundant queries, resulting in fast execution and concise results. Furthermore, using a novel technique Conqueror is capable of detecting previously unknown functional dependencies that hold on the database relations as well as on joins of relations. These newly detected dependencies are then used to prune even more redundant queries. Conqueror has an efficient database-oriented implementation using SQL.
Conqueror+ Java source code by Wim Le Page
Mining Association Rules of Simple Conjunctive Queries by Bart Goethals, Wim Le Page, and Heikki Mannila in Proc. SIAM Int. Conf. on Data Mining
IMDB snapshot dataset (available upon request)
QuizDB dataset (available upon request)