Top-k Approximate Answers to XPath Queries with Negation

Data heterogeneity in XML retrieval activities can be tackled by giving users the possibility to obtain approximate answers to their queries. XPath query relaxation has been proposed as a mechanism to provide approximate answers in the case of positive XPath queries. Under this mechanism, the "satisfaction score" of an answer is defined by looking at how the query must be relaxed to produce that answer, and the user is provided with the best k answers according to their satisfaction score (top-k query answering). In this paper we investigate the problem of top-k query answering for XPath queries with negation. We tackle the challenging issues that need to be carefully considered when dealing with the approximation of negated conditions and propose an incremental top-k query answering technique based on query relaxation. Specifically, after defining a weighted query language and its semantics, we develop a general incremental query evaluation framework, which is flexible enough to support different evaluation strategies. The experimental assessment confirms the effectiveness of the whole framework.