Sanja Štajner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Sanja Štajner has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
Simone Paolo Ponzetto
6 shared publications- Heiner Stuckenschmidt
University of Mannheim
4 shared publications - Horacio Saggion
Universitat Pompeu Fabra
3 shared publications Goran Glavaš
2 shared publicationsHoracio Saggion
Universitat Pompeu Fabra, Barcelona, Spain
2 shared publications- Seren Yenikent
Symanto Research GmbH & Co KG, Nuernberg, Germany
2 shared publications
Latest Publications
- Open Access
Lexical Simplification Benchmarks for English, Portuguese, and Spanish
Even in highly-developed countries, as many as 15-30\% of the population can only understand texts written using a basic vocabulary. Their understanding of everyday texts is limited, which prevents them from taking an active role in society and making informed decisions regarding healthcare, legal representation, or democratic choice. Lexical simplification is a natural language processing task that aims to make text understandable to everyone by replacing complex vocabulary and expressions with simpler ones, while preserving the original meaning. It has attracted considerable attention in the last 20 years, and fully automatic lexical simplification systems have been proposed for various languages. The main obstacle for the progress of the field is the absence of high-quality datasets for building and evaluating lexical simplification systems. We present a new benchmark dataset for lexical simplification in English, Spanish, and (Brazilian) Portuguese, and provide details about data selection and annotation procedures. This is the first dataset that offers a direct comparison of lexical simplification systems for three languages. To showcase the usability of the dataset, we adapt two state-of-the-art lexical simplification systems with differing architectures (neural vs.\ non-neural) to all three languages (English, Spanish, and Brazilian Portuguese) and evaluate their performances on our new dataset. For a fairer comparison, we use several evaluation measures which capture varied aspects of the systems' efficacy, and discuss their strengths and weaknesses. We find a state-of-the-art neural lexical simplification system outperforms a state-of-the-art non-neural lexical simplification system in all three languages. More importantly, we find that the state-of-the-art neural lexical simplification systems perform significantly better for English than for Spanish and Portuguese.
Open AccessLexical Simplification Benchmarks for English, Portuguese, and Spanish
Even in highly-developed countries, as many as 15-30\% of the population can only understand texts written using a basic vocabulary. Their understanding of everyday texts is limited, which prevents them from taking an active role in society and making informed decisions regarding healthcare, legal representation, or democratic choice. Lexical simplification is a natural language processing task that aims to make text understandable to everyone by replacing complex vocabulary and expressions with simpler ones, while preserving the original meaning. It has attracted considerable attention in the last 20 years, and fully automatic lexical simplification systems have been proposed for various languages. The main obstacle for the progress of the field is the absence of high-quality datasets for building and evaluating lexical simplification systems. We present a new benchmark dataset for lexical simplification in English, Spanish, and (Brazilian) Portuguese, and provide details about data selection and annotation procedures. This is the first dataset that offers a direct comparison of lexical simplification systems for three languages. To showcase the usability of the dataset, we adapt two state-of-the-art lexical simplification systems with differing architectures (neural vs.\ non-neural) to all three languages (English, Spanish, and Brazilian Portuguese) and evaluate their performances on our new dataset. For a fairer comparison, we use several evaluation measures which capture varied aspects of the systems' efficacy, and discuss their strengths and weaknesses. We find a state-of-the-art neural lexical simplification system outperforms a state-of-the-art non-neural lexical simplification system in all three languages. More importantly, we find that the state-of-the-art neural lexical simplification systems perform significantly better for English than for Spanish and Portuguese.
- 1 March 2019
- Published by Elsevier in Expert Systems with Applications
- Vol. 118, 80-91
Improving lexical coverage of text simplification systems for Spanish
- 1 March 2019
- Published by Elsevier in Expert Systems with Applications
- Vol. 118, 80-91
Improving lexical coverage of text simplification systems for Spanish
- 10 October 2018
- Published by Springer Nature
- p. 642-654
Automatic Detection of Uncertain Statements in the Financial Domain
- 10 October 2018
- Published by Springer Nature
- p. 642-654
Automatic Detection of Uncertain Statements in the Financial Domain
- 10 November 2017
- Published by Assoc. for Computational Linguistics Bulgaria
Multilingual and Cross-Lingual ComplexWord Identification
- 10 November 2017
- Published by Assoc. for Computational Linguistics Bulgaria
Multilingual and Cross-Lingual ComplexWord Identification
- 1 October 2017
- Published by Elsevier in Expert Systems with Applications
- Vol. 82, 383-395
Leveraging event-based semantics for automated text simplification
- 1 October 2017
- Published by Elsevier in Expert Systems with Applications
- Vol. 82, 383-395
Leveraging event-based semantics for automated text simplification
- 1 August 2017
- Published by International Joint Conferences on Artificial Intelligence
- p. 4096-4102
Automatic Assessment of Absolute Sentence Complexity
Lexically and syntactically simpler sentences result in shorter reading time and better understanding in many people. However, no reliable systems for automatic assessment of absolute sentence complexity have been proposed so far. Instead, the assessment is usually done manually, requiring expert human annotators. To address this problem, we first define the sentence complexity assessment as a five-level classification task, and build a ‘gold standard’ dataset. Next, we propose robust systems for sentence complexity assessment, using a novel set of features based on leveraging lexical properties of freely available corpora, and investigate the impact of the feature type and corpus size on the classification performance.
- 1 August 2017
- Published by International Joint Conferences on Artificial Intelligence
- p. 4096-4102
Automatic Assessment of Absolute Sentence Complexity
Lexically and syntactically simpler sentences result in shorter reading time and better understanding in many people. However, no reliable systems for automatic assessment of absolute sentence complexity have been proposed so far. Instead, the assessment is usually done manually, requiring expert human annotators. To address this problem, we first define the sentence complexity assessment as a five-level classification task, and build a ‘gold standard’ dataset. Next, we propose robust systems for sentence complexity assessment, using a novel set of features based on leveraging lexical properties of freely available corpora, and investigate the impact of the feature type and corpus size on the classification performance.
- Open Access
- 1 January 2017
- Published by Association for Computational Linguistics (ACL)
Exploring Neural Text Simplification Models
- 1 January 2017
- Published by Association for Computational Linguistics (ACL)
Open AccessExploring Neural Text Simplification Models
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 164-171
Domain Adaptation for Automatic Detection of Speculative Sentences
The use of speculative, uncertain or vague sentences is common in both spoken and written language. Automatic detection of such sentences plays significant role in natural language processing and social sciences as it could enhance information extraction systems and lead to faster and more reliable analyses in social sciences. However, this problem has only been addressed in biomedical and encyclopedic domains. In this paper, we address automatic speculation detection (as a binary sentence classification task) in monetary policy domain, and for the first time, on the transcripts of spoken language. We build two new speculation detection datasets and a dictionary of speculation triggers using expert annotations, and benchmark the performance of automatic speculation detection systems in this new domain.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 164-171
Domain Adaptation for Automatic Detection of Speculative Sentences
The use of speculative, uncertain or vague sentences is common in both spoken and written language. Automatic detection of such sentences plays significant role in natural language processing and social sciences as it could enhance information extraction systems and lead to faster and more reliable analyses in social sciences. However, this problem has only been addressed in biomedical and encyclopedic domains. In this paper, we address automatic speculation detection (as a binary sentence classification task) in monetary policy domain, and for the first time, on the transcripts of spoken language. We build two new speculation detection datasets and a dictionary of speculation triggers using expert annotations, and benchmark the performance of automatic speculation detection systems in this new domain.
- 21 June 2016
- Published by Springer Nature
- p. 50-61
Domain-Specific Hybrid Machine Translation from English to Portuguese
- 21 June 2016
- Published by Springer Nature
- p. 50-61
Domain-Specific Hybrid Machine Translation from English to Portuguese