Proceedings of the Workshop on Automatic Text Simplification - Methods and Applications in the Multilingual Society (ATS-MA 201 2014
DOI: 10.3115/v1/w14-5601
|View full text |Cite
|
Sign up to set email alerts
|

Relative clause extraction for syntactic simplification

Abstract: This paper investigates non-destructive simplification, a type of syntactic text simplification which focuses on extracting embedded clauses from structurally complex sentences and rephrasing them without affecting their original meaning. This process reduces the average sentence length and complexity to make text simpler. Although relevant for human readers with low reading skills or language disabilities, the process has direct applications in NLP. In this paper we analyse the extraction of relative clauses … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
8
0

Year Published

2015
2015
2021
2021

Publication Types

Select...
3
2
1

Relationship

0
6

Authors

Journals

citations
Cited by 7 publications
(9 citation statements)
references
References 13 publications
1
8
0
Order By: Relevance
“…Each row specifies a different feature set -"All" specifies the entire feature set from Table 2, while each subsequent line removes one type of features. (Dornescu et al, 2014) performs better on our dataset Their method achieves much better results on our dataset (compare 64% overall F1 on our dataset with their reported 45.29% F1 on their dataset). This speaks both for their method as a valid signal for restrictiveness annotation, as well as for the improved consistency of our dataset.…”
Section: Gold Treessupporting
confidence: 49%
See 1 more Smart Citation
“…Each row specifies a different feature set -"All" specifies the entire feature set from Table 2, while each subsequent line removes one type of features. (Dornescu et al, 2014) performs better on our dataset Their method achieves much better results on our dataset (compare 64% overall F1 on our dataset with their reported 45.29% F1 on their dataset). This speaks both for their method as a valid signal for restrictiveness annotation, as well as for the improved consistency of our dataset.…”
Section: Gold Treessupporting
confidence: 49%
“…Replicating (Dornescu et al, 2014) Their best performing ML-based algorithm 10 uses the supervised CRFsuite classifier (Okazaki, 2007) over "standard features used in chunking, such as word form, lemma and part of speech tags". Replicating their baseline, we extract the list of features detailed in Table 2 (in the row labeled "chunking features").…”
Section: Baselinesmentioning
confidence: 99%
“…Relative clauses are one of the most features clauses in constructing complex sentences. Dornescu et al [15] have specifically worked on Relative clause extraction for syntactic simplification. Their process reduced the average sentence length and complexity to make text simpler.…”
Section: Previous Workmentioning
confidence: 99%
“…Inspired by the work of Aranzabe et al [9], Sharma et al [14] and Dornescu et al [15], we have targeted relative clauses present in the complex sentences and provide a better approach to identify and extract them for simplification process.…”
Section: Previous Workmentioning
confidence: 99%
“…Early researches in text simplification have considered punctuation marks such as comma, semicolon, and parentheses as important markers to split the clauses [19]. An extensive research of relative clause extraction on simplifying text is described in [20].…”
Section: Introductionmentioning
confidence: 99%