Recursive query optimization and partial evaluation

The paper presents an approach to recursive query optimization based upon the integration of partial evaluation and already existing rule transformation based, optimization methods such as Magic Set, Minimagic and Counting. The basic idea is to partially evaluate a logic program with respect to a query (goal), thus eliminating unnecessary intermediate (IDB) predicates and then further optimize the program, with respect to evaluation, by using one of the mentioned optimization methods. The advantage of the proposal is twofold: on one hand to give the user freedom of using all the power of a logical language to define the query program and, on the other, to keep efficiency into reasonable range.