Applying machine learning techniques to automating XML markup

In this paper we present a novel system which uses machine-learning techniques to automatically markup text documents into XML. The system uses the WEBSOM method to arrange marked-up documents on a Self-Organizing Map (SOM) such that semantically similar documents are placed close to each other. Then, by using the inductive learning algorithm, C5.0, the system automatically extracts and applies markup rules from the nearest SOM neighbours of an unmarked document. The system is designed to be adaptive, so that once a document is marked-up, its behaviour is modified to improve accuracy. Our approach has been applied to a number of different document domains and the results indicate that the approach is feasible and practical.