Supporting Structural and Keyword Search for XML Data

• Main Authors: Chieh-Chang Luo, 羅介璋
• Other Authors: Ya-Hui Chang
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/85328509214377290085

碩士 === 國立臺灣海洋大學 === 資訊工程學系 === 94 === Abstract In this thesis, we discuss how to use the technology of Information Retrieval to combine with the XQuery Query Processing algorithm, so that we could retrieve the data that conforms the structural and value constraints imposed by users. To process the structural constraints, we propose to decompose the query tree into a sequence of suffix paths. Each suffix path consists of only parent-child relationship between nodes. Then, the system will retrieve elements that correspond to each suffix path, and combine those elements to form the final result. The XML data will be pre-processed and represented in the internal form. Basically, we assign each possible path in XML data a P-Label. Each path is associated with an Element-Encoding table, which records the structural and value information for all the elements which correspond to this P-Label. To meet the requirement of information retrieval, we adopt the BM25 Scoring Model to pre-compute the scores between elements and keywords and store the scores within the associated Element-Encoding Table. Therefore, we can efficiently determine if the element’s keyword satisfies user’s value constraint and find the corresponding score. We have performed a series of experiments to evaluate the performance of the proposed approach. We also contrast with a well-known system. The experimental results show that our system is very efficient.