Constraint-Based Synthesis of Datalog Programs

Data migration using datalog program synthesis

“…Dynamite builds the target database instance from the output facts of the synthesized Datalog program as described in Section 3. 3. However, Dynamite performs one optimization to make largescale data migration practical: We leverage MongoDB [2] to build indices on attributes that connect records to their parents.…”

“…Among ILP techniques, our work is most similar to recent work on Datalog program synthesis [3,42]. In particular, Zaatar [3] encodes an under-approximation of Datalog semantics using the theory of arrays and reduces synthesis to SMT solving. However, this technique imposes an upper bound on the number of clauses and atoms in the Datalog program.…”

“…To answer RQ4, we compare Dynamite against a baseline called Dynamite-Enum that uses enumerative search instead of the sketch completion technique described in Section 4. 3. In particular, Dynamite-Enum uses the lazy enumeration algorithm based on SMT, but it does not use the Analyze procedure for learning from failed synthesis attempts.…”

SLD-Resolution Reduction of Second-Order Horn Fragments

“…These clauses are such that C 1 , C 2 ∈ H c a ,∞ and C 1 , C 2 C modulo variable unification. 1 Thus, C is reducible. 1 Some connections may be lost between variables in C1 and C2 since only the ones occurring in the spanning tree S are preserved.…”

“…1 Thus, C is reducible. 1 Some connections may be lost between variables in C1 and C2 since only the ones occurring in the spanning tree S are preserved. However, they can be recovered by unifying the disconnected variables together in the resolvent.…”