Jumping over useless nodes in Xml tree pattern with Level Based Index

XML queries are based on path expressions which are equal to a combination of some elements connected to each other in a tree pattern structure, called Tree Pattern Query (TPQ). Therefore, the main operation in XML query processing is to find the nodes that match the given QTP in the document. A number of methods are offered for QTP matching, but the majority of these methods for accessing to all QTPs in the document, process all of the query nodes, So that they are very time-consuming. A few methods, such as TJFast, process only nodes which satisfy the leaves of QTP, so these methods need too less node access. But these methods also have to access too many useless nodes. Useless nodes are those that either do not participate in final result or produce repetitive result if they are processed. In this paper, by presenting Level-Based Index (LBI), we propose a method to index query leaves. By use of this structure, we could jump over useless nodes.