An Efficient Algorithm for XML Keyword Search Based on Interval Encoding

Recently, keyword search on XML data has received much attention. Existing XML keyword search algorithms are all based on Dewey labeling, and this method in the calculation of the common ancestor would suffer from the CAR (common-ancestor-repetition) problem. In this paper, we propose a novel index based on interval encoding, namely IS-Index. We then design an efficient algorithm based on this structure, namely Interval-Search algorithm. Firstly, the algorithm calculates the minimum common ancestors by the range value between nodes that contain keywords. Then it traverses the IS-Index and computes common ancestor nodes to get the candidate SLCAs. Finally, it filters the candidate SLCAs to get SLCA nodes. The experimental results show that our method is an efficient algorithm for finding SLCA nodes.