A Semantic Framework for Summarizing XML Documents

EXtensible Markup Language (XML) represents data in an efficient way due to its flexibility and the availability to use in various applications. The need to summarize XML documents is increased due to the increasing use of XML in data exchange and representation also due to its difficulty to read and understand. This paper presents an XML Abstractive Summary (XAS) approach to summarize the XML document in a semantic and concise way. The experiments are done using two dataset: IMDB and DBLP. The results has been tested with more than 300000 XML documents. XAS approach decreases the size of the document up to 50 % with average precision and recall 76.5% and 45% respectively.