Evaluating Syntactic Sentence Compression for Text Summarisation

Abstract

Abstract. This paper presents our work on the evaluation of syntactic based sentence compression for automatic text summarization. Sentence compression techniques can contribute to text summarization by removing redundant and irrelevant information and allowing more space for more relevant content. However, very little work has focused on evaluating the contribution of this idea for summarization. In this paper, we focus on pruning individual sentences in extractive summaries using phrase structure grammar representations. We have implemented several syntaxbased pruning techniques and evaluated them in the context of automatic summarization, using standard evaluation metrics. We have performed our evaluation on the TAC and DUC corpora using the BlogSum and MEAD summarizers. The results show that sentence pruning can achieve compression rates as low as 60%, however when using this extra space to fill in more sentences, ROUGE scores do not improve significantly.

Trends in Multi-document Summarization System Methods

“…Compression before summarisation is very important as it allows guide in selecting sentences without considering adjuncts in the sentences. The work of [29] focused on evaluating automatic summary based on syntactic pruning of sentences. Mead and Blogsum [1] summarizers were used to test the method.…”

“…From the papers reviewed on sentence reduction, it can be regarded as abstractive summary since the sentences present are not exactly what is in the document. For example, sentence in [29] review will be…”

“…Also, it is deduced that syntactic based compression does not improve a generic summarization system [27,28,29,31] when evaluated by state-of-the-art standard but can achieve better performance if semantic role information can be incorporated into the model.…”

A Sentence Simplification System for Improving Relation Extraction

“…Referring to previous attempts at syntax-based sentence compression (Dunlavy et al, 2003;Zajic et al, 2007;Perera and Kosseim, 2013), the idea of our text simplification framework is to syntactically simplify a complex input sentence by splitting conjoined clauses into separate sentences and by eliminating specific syntactic sub-structures, namely those containing only minor information. However, unlike recent approaches in the field of extractive sentence compression, we do not delete these constituents, which would result in a loss of background information, but rather aim at preserving the full informational content of the original sentence.…”