A tree parent storage based on hashtable for XML construction

The emergence of the Web has increased interests in XML data. Labeling methods to extract and reconstruct the structural information of XML data, which are important for many applications such as XML IR, are becoming more attractive. To achieve efficient structural information extraction, in this paper we propose a tree parent storage based on hashtable, a novel dynamic-friendly encoding scheme, which has great advantages both in storage space and time of constructing label paths compared with prefix schema. We have conducted an experimental study on real-life XML data sets, on Berkeley DB to evaluate and compare the performance and storage consumption of the proposed method with prefix schema. The experimental results show that our schema save about 60%-70% space in structural information storing whereas having the same performance in executing time for a certain query compared with prefix schema. In conclusion,it is effective, and efficient when compared with existing proposals.