Summarization of Abstractive Multi-document using Sub-graph & Network

Abstract

Automatic multi-document theoretical summarization system is employed to summarize many documents
into a brief one with generated new sentences. Several of them square measure supported word-graph and ILP technique,
and much of sentences square measure unnoticed thanks to the serious computation load. to cut back computation and
generate legible and informative summaries, we have a tendency to propose a completely unique theoretical multidocument summarization system supported chunk-graph (CG) and repeated neural network language model (RNNLM).
In our approach, A CG that relies on word-graph is made to prepare all data in a very sentence cluster, CG will cut back
the scale of graph and keep additional linguistics data than word-graph. we have a tendency to use beam search and
character-level RNNLM to come up with legible and informative summaries from the CG for every sentence cluster,
RNNLM may be a higher model to gauge sentence linguistic quality than n-gram language model. Experimental results
show that our planned system outperforms all baseline systems and reach the state-of-art systems, and therefore the
system with CG will generate higher summaries than that with standard word-graph.