The Effect of Various Straightforward Summarization Techniques on Informativeness of Abstractive Summaries


In the present paper we describe how a straight forward abstractive text summarization method based on syntactic text simplification, subject-verb-object concept frequency scoring and a set of rules that transform text in to its semantic representation improves informativeness of automatic summaries. We analyze the impact of each individual component of our approach on the quality of generated summaries and test it on DUC 2002 dataset. Our experiments show that our approach outperforms other state-of-the-art abstractive methods while maintaining acceptable linguistic quality and redundancy rate.

Revista: RANLP 2019

Autores: Tatiana Vodolazova and Elena Lloret