Top-Down XML Keyword Query Processing

Abstract

XML keyword queries have been very effective in the last decade. The main factors resulting from the ineffectiveness of current methods are common problems of repetition of progress (CAR) and non-beneficial contractual problems (VUN). To address the CAR problem, we suggest a top-down processing strategy to answer the selected keyword query w.r.t. Significance of LCA / SLCA / ELCA. We want to say "from the top down" that we visit all CA nodes for the first time, from left to right; through "general", we mean that our method is independent of query semantics. To address the VUN problem, we suggest that you use secondary nodes, rather than secondary nodes, to test the satis capacity on node v w.r.t. Specific indicators We suggest an algorithm based on traditional inverted lists or recently proposed lists of LLists to improve overall performance. We also propose several search-based algorithms to simplify running CA nodes for all relevant LLists lists. Experimental results verify the benefits of our methods according to different evaluation measures.