Exploiting Cross-Document Relations for Multi-document Evolving Summarization

This paper presents a methodology for summarization from multiple documents which are about a specic topic. It is based on the specication and identication of the cross-document relations that occur among textual elements within those documents. Our methodology involves the specication of the topic-specic entities, the messages conveyed for the specic entities by certain textual elements and the specication of the relations that can hold among these messages. The above resources are necessary for setting up a specic topic for our query-based summarization system which uses these resources to identify the queryspecic messages within the documents and the query-specic relations that connect these messages across documents.

• Type: Conference
• Year: 2004
• Where: SETN
• Authors: Stergos D. Afantenos, Irene Doura, Eleni Kapellou, Vangelis Karkaletsis
• Keywords: Artificial Intelligence, Certain Textual Elements, SETN 2004, Specic Topic, Textual Elements