A System for Keyword Search on Probability XML Data

摘要

Many probabilistic XML data models have been proposed to store XML data with uncertainty information, and based on them the issues such as structured querying are extensively studied. As an alternative to structured querying, keyword search in probabilistic XML data needs to be concerned. In this paper we addressed the issue of keyword search on probabilistic XML data. The probabilistic XML data is viewed as a labeled tree, and a concept of Minimum Meaningful Fragment (MMF) is defined as the searching result. A MMF is a minimum subtree of the probabilistic XML data which has a positive probability of containing all keywords. To sort the MMFs a novel scoring function mainly considering the degree of uncertainty information is presented. We propose a system to compute top-k searching results efficiently based on the scoring function. The experiments shows the efficiency for our system.