Abstract: The emergence of the Web has increased interests in XML data. Keyword search has attracted a great deal of attention for retrieving XMI_ data because it is a user-friendly mechanism. 13ut Keyword search is hard to directly im- prove search quality because lots of keyword-matched nodes may not contribute to the results. A more important issue is the current studies are focused on single XML retrieval, lack of practicability. To address the challenge, this article pro- posed a new approach for automatically correcting queries over Multi-XMI，called MXDR(Multi XML Distributed Re- trieval). We first classed multi XML documents by a clustering method,and elicited the common structure information. Then generated certifiable structured queries by analyzing the given keywords query and the common structure informa- tion of XML datasets. We can evaluate the generated structured queries over the XML data sources with any existing structure search engine. We conducted an experimental study on real-life multi-XML datasets. The experimental results show that MXDR is effective and efficient in supporting structural querics,compared with existing proposals.

Key words: Multi-ML，Keywords IR, Structure IR, Distributed