A Keywords Search Method over Relational Databases Based on Join Tree

A solution based on join tree is proposed to improve the efficiency of keywords search over relational databases (KSORD). In the solution, a join tree and a select template are generated from a schema graph. During a query of KSORD, select template of the candidate join tree can be found out efficiently, instead of finding it from a schema graph by Breadth-First Search. Finally, an experiment is verified that the solution is more efficiently than the current KSORD.