Title :
Generating user-focused, content-based summaries for multi-documents using document graphs
Author :
Mohamed, Ahmed A.
Author_Institution :
Dept. of Comput. Sci. & Eng., Connecticut Univ., Storrs, CT
Abstract :
Due to the huge amount of information available on-line, the research and development of systems to automatically summarize multiple documents has become the focus of considerable interest and investment in many research sectors. One of the challenges that face multi-document summarization is generating user-focused summaries. Here, we present a new method for automatic user-focused summarization generation of multiple documents using document graphs. We propose generating summaries based on identifying the important concepts/entities and relations in the text. In this approach we are generating 3 different summaries, each one of them belongs to a different level of focus. The data from Document Understanding Conference 2002 (DUC2002) has been used in the experiment
Keywords :
content management; research and development; text analysis; content-based summaries; document graphs; multidocument summarization; research and development; user-focused content-based summaries; Biomedical signal processing; Books; Computer science; Data mining; Information analysis; Investments; NIST; Research and development; Signal processing algorithms; Writing;
Conference_Titel :
Signal Processing and Information Technology, 2005. Proceedings of the Fifth IEEE International Symposium on
Conference_Location :
Athens
Print_ISBN :
0-7803-9313-9
DOI :
10.1109/ISSPIT.2005.1577178