MultiSum : query-based multi-document summarisation

Abstract

The explosion of digital information and the penetration of the Internet has given rise to the problem of Information Overload. This final year project describes an open-domain Multi-Document Summarisation architecture which is used to build a system capable of automatically identifying query-related online documents and compiling a report from the most relevant sources, whilst presenting the result in such a way as to make it easy for the researcher to look up the information in its original context, maintain the report and also publish it to external systems.