FVC: A Feature-Vector-Based Classification for XML Dissemination

With the adoption of XML in a wide range of applications, efficient XML classification has become an important research topic. In current studies, users' interests are expressed by XPath or XQuery queries. However, such a query is hard to formulate, because it requires a good knowledge of the structure and contents of the documents that will arrive and some knowledge of XQuery which few consumers will have. The query may even be impossible to formulate in cases where the distinction of relevant and irrelevant documents requires the consideration of a large number of features. Traditional classification method can't work well for XML dissemination, because the number of training example is often small. Therefore, this paper introduces a data mining approach to XML dissemination that uses a given document collection of the user to automatically learn a classifier modelling his/her information needs. We present a novel XML classifier taking into account the structure as well as the content of XML documents. Our experimental evaluation on several real XML document sets demonstrates the accuracy and efficiency of the proposed XML classification approach.