Automatic Text Simplification for German

Bibliographic Data

ID22093290
AuthorsSarah Ebling (0000-0001-6511-5085, University of Zurich, corresponding author), Alessia Battisti (University of Zurich), Marek Kostrzewa (0000-0002-1352-4546, University of Zurich), Dominik Pfütze (University of Zurich), Annette Rios (0000-0002-8943-3472, University of Zurich), Andreas Säuberli (0000-0001-9613-334X, University of Zurich), Nicolas Spring (0000-0001-6247-7646, University of Zurich)
Year2022
Volume7
Publication date2022-02-23
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueFrontiers in Communication (JOURNAL)
Journal identifiersISSN: 2297-900X • E-ISSN: 2297-900X
PublisherFrontiers Media SA (PUBLISHER • CH)
DOI10.3389/fcomm.2022.706718
OpenAlexW4214730182
LanguageEN
Citations received1
References cited18

The article at hand aggregates the work of our group in automatic processing of simplified German. We present four parallel (standard/simplified German) corpora compiled and curated by our group. We report on the creation of a gold standard of sentence alignments from the four sources for evaluating automatic alignment methods on this gold standard. We show that one of the alignment methods performs best on the majority of the data sources. We used two of our corpora as a basis for the first sentence-based neural machine translation (NMT) approach toward automatic simplification of German. In follow-up work, we extended our model to render it capable of explicitly operating on multiple levels of simplified German. We show that using source-side language level labels improves performance with regard to two evaluation metrics commonly applied to measuring the quality of automatic text simplification

German · Linguistics · Machine translation · Natural language processing · Sentence · Training set · Computer Science · Mathematics · Natural Language Processing Techniques · Text Readability and Simplification · Topic Modeling · Artificial Intelligence

Tools

Open DOIOpen Access