Hybrid text simplification using synchronous dependency grammars with hand-written and automatically harvested rules

Siddharthan, Advaith; Mandya, Angrosh

doi:10.3115/v1/e14-1076

Cited by 53 publications

(59 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 3 shows the results for the complete data set. As can be observed, our system achieved the same mean score for simplicity as Siddharthan and Angrosh's [25], and is slightly better at fluency and adequacy, though not statistically significant. Both automatic systems are comparable to the Simple Wikipedia version.…”

Section: Intrinsic Evaluation Of the Yats Syntactic Simplification Cosupporting

confidence: 62%

“…Coster and Kauchak [10] and Wubben et al [28] viewed simplification as a monolingual translation task and they applied phrase based MT to the text simplification augmented with a phrasal deletion model [10] and a post-hoc reranking procedure that ranked the output [28]. Siddharthan and Angrosh [25] presented a hybrid system that combined manually written synchronous grammars for syntactic simplificatins with an automatically acquired syncronous grammar for lexicalised constructs.…”

Section: Related Workmentioning

confidence: 99%

“…Participants were presented with the source sentence from the English Wikipedia followed by three simplified versions from Simple Wikipedia (SW) (i.e. a sentence from Simple Wikipedia aligned to the sentence in English Wikipedia), the system developed by Siddharthan and Angrosh [25] (S&A), and YATS, in a randomisd order, and they were asked to rate each of the simplified version w.r.t. the extend to which it was grammatical (fluency), the extend to which it had the same meaning as the complex sentence (adequacy), and the extend to which it was simpler than the complex and thus easier to understand (simplicity).…”

Section: Intrinsic Evaluation Of the Yats Syntactic Simplification Comentioning

confidence: 99%

“…We performed manual evaluation relying on eight human judges, 12 who assessed our system w.r.t. fluency, adequacy, and simplicity, using the evaluation set used by Siddharthan and Angrosh [25], from which we randomly selected 25 sentences. Participants were presented with the source sentence from the English Wikipedia followed by three simplified versions from Simple Wikipedia (SW) (i.e.…”

Section: Intrinsic Evaluation Of the Yats Syntactic Simplification Comentioning

confidence: 99%

See 3 more Smart Citations

YATS: Yet Another Text Simplifier

Ferrés

Marimon

Saggion

et al. 2016

Natural Language Processing and Information Systems

View full text Add to dashboard Cite

Abstract. We present a text simplifier for English that has been built with open source software and has both lexical and syntactic simplification capabilities. The lexical simplifier uses a vector space model approach to obtain the most appropriate sense of a given word in a given context and word frequency simplicity measures to rank synonyms. The syntactic simplifier uses linguistically-motivated rule-based syntactic analysis and generation techniques that rely on part-of-speech tags and syntactic dependency information. Experimental results show good performance of the lexical simplification component when compared to a hard-to-beat baseline, good syntactic simplification accuracy, and according to human assessment, improvements over the best reported results in the literature for a system with same architecture as YATS.

show abstract

Section: Intrinsic Evaluation Of the Yats Syntactic Simplification Cosupporting

confidence: 62%

Section: Related Workmentioning

confidence: 99%

Section: Intrinsic Evaluation Of the Yats Syntactic Simplification Comentioning

confidence: 99%

Section: Intrinsic Evaluation Of the Yats Syntactic Simplification Comentioning

confidence: 99%

See 2 more Smart Citations

YATS: Yet Another Text Simplifier

Ferrés

Marimon

Saggion

et al. 2016

Natural Language Processing and Information Systems

View full text Add to dashboard Cite

show abstract

“…For example, systems which allow for some global optimization, such as integer linear programming. They use synchronous grammar that combines a manually constructed grammar for syntactic rules and an automatically acquired grammar for lexical rules and paraphrase [20], [44], [45], [46]. This analysis could be extended to other systems mentioned in Section 2.…”

Section: The Model As An Abstraction Of Existing Practicementioning

confidence: 99%

An approach to treat numerical information in the text simplification process

Bautista

Hervás

Gervás

et al. 2015

Univ Access Inf Soc

View full text Add to dashboard Cite

Abstract:Public information services and documents should be accessible to the widest possible readership. In particular, information from these sources often takes the form of numerical expressions, which pose comprehension problems for many people, including people with disabilities, who are often also exposed to poverty, illiteracy, or lack of access to advanced technology. This paper presents an approach to treat numerical information in the text simplification process to make it more accessible. A generic model for automatic text simplification systems is presented, aimed at making documents more accessible to readers with cognitive disabilities. The proposed approach is validated with a real system to simplify numerical expressions in Spanish. This system is then evaluated and the results show that it is appropriate for the task at hand.

show abstract

Sentence simplification, compression, and disaggregation for summarization of sophisticated documents

Finegan-Dollak

Radev

2015

Asso for Info Science & Tech

View full text Add to dashboard Cite

Sophisticated documents like legal cases and biomedical articles can contain unusually long sentences. Extractive summarizers can select such sentences—potentially adding hundreds of unnecessary words to the summary—or exclude them and lose important content. Sentence simplification or compression seems on the surface to be a promising solution. However, compression removes words before the selection algorithm can use them, and simplification generates sentences that may be ambiguous in an extractive summary. We therefore compare the performance of an extractive summarizer selecting from the sentences of the original document with that of the summarizer selecting from sentences shortened in three ways: simplification, compression, and disaggregation, which splits one sentence into several according to rules designed to keep all meaning. We find that on legal cases and biomedical articles, these shortening methods generate ungrammatical output. Human evaluators performed an extrinsic evaluation consisting of comprehension questions about the summaries. Evaluators given compressed, simplified, or disaggregated versions of the summaries answered fewer questions correctly than did those given summaries with unaltered sentences. Error analysis suggests 2 causes: Altered sentences sometimes interact with the sentence selection algorithm, and alterations to sentences sometimes obscure information in the summary. We discuss future work to alleviate these problems.

show abstract

Hybrid text simplification using synchronous dependency grammars with hand-written and automatically harvested rules

Cited by 53 publications

References 16 publications

YATS: Yet Another Text Simplifier

YATS: Yet Another Text Simplifier

An approach to treat numerical information in the text simplification process

Sentence simplification, compression, and disaggregation for summarization of sophisticated documents

Contact Info

Product

Resources

About