Putting a Spotlight on Validators of Easy-to-Read Content

Easy-to-Read content results from applying text simplification principles to make information accessible for persons with reading and learning difficulties. While both the creation process and simplification principles have gained the interest of scholars and the general public in the past years, the role of validators is still less visible compared with that of writers or translators. This paper sought to put a spotlight on validators by answering the questions of who these professionals are, what tasks they take on, and how they have acquired the necessary knowledge and skills for the job. In doing so, it investigated a subset of the data about validators’ demographical and educational backgrounds and current activity collected in an online survey launched within the innovative framework of the Erasmus+ project Train2Validate.

Approaches to Text Simplification: Can Computer Technologies Outdo a Human Mind?

Narrowly specialized information is addressed to a limited circle of professionals though it provokes interest among people without specialized education. This gives rise to a need for the popularization of scientific information. This process is carried out through simplified texts as a kind of secondary texts that are directly aimed at the addressee. Age, language proficiency and background knowledge are the main features which are usually taken into consideration by the author of the secondary text who makes changes in the text composition, as well as in its pragmatics, semantics and syntax. This article analyses traditional approaches to text simplification, computer simplification and summarization. The authors compare human-authored simplification of literary texts with the newest trends in computer simplification to promote further development of machine simplification tools. It has been found that the samples of simplified scientific texts seem to be more natural than the samples of simplified literary texts since technical background knowledge can be processed with machine tools. The authors have come to the conclusion that literary and technical texts should imply different approaches for adaptation and simplification. In addition, personal readers’ experience plays a great part in finding the implications in literary texts. In this respect it might be reasonable to create separate engines for simplifying and adapting texts from diverse spheres of knowledge. Keywords Text Simplification; Natural Language Processing (NLP); Pragmatic Adaptation; Professional Communication; Literary Texts

The (Un)Suitability of Automatic Evaluation Metrics for Text Simplification

In order to simplify sentences, several rewriting operations can be performed such as replacing complex words per simpler synonyms, deleting unnecessary information, and splitting long sentences. Despite this multi-operation nature, evaluation of automatic simplification systems relies on metrics that moderately correlate with human judgements on the simplicity achieved by executing specific operations (e.g. simplicity gain based on lexical replacements). In this article, we investigate how well existing metrics can assess sentence-level simplifications where multiple operations may have been applied and which, therefore, require more general simplicity judgements. For that, we first collect a new and more reliable dataset for evaluating the correlation of metrics and human judgements of overall simplicity. Second, we conduct the first meta-evaluation of automatic metrics in Text Simplification, using our new dataset (and other existing data) to analyse the variation of the correlation between metrics’ scores and human judgements across three dimensions: the perceived simplicity level, the system type and the set of references used for computation. We show that these three aspects affect the correlations and, in particular, highlight the limitations of commonly-used operation-specific metrics. Finally, based on our findings, we propose a set of recommendations for automatic evaluation of multi-operation simplifications, suggesting which metrics to compute and how to interpret their scores.

ALaSca: an Automated approach for Large-Scale Lexical Substitution

The lexical substitution task aims at finding suitable replacements for words in context. It has proved to be useful in several areas, such as word sense induction and text simplification, as well as in more practical applications such as writing-assistant tools. However, the paucity of annotated data has forced researchers to apply mainly unsupervised approaches, limiting the applicability of large pre-trained models and thus hampering the potential benefits of supervised approaches to the task. In this paper, we mitigate this issue by proposing ALaSca, a novel approach to automatically creating large-scale datasets for English lexical substitution. ALaSca allows examples to be produced for potentially any word in a language vocabulary and to cover most of the meanings it lists. Thanks to this, we can unleash the full potential of neural architectures and finetune them on the lexical substitution task. Indeed, when using our data, a transformer-based model performs substantially better than when using manually annotated data only. We release ALaSca at https://sapienzanlp.github.io/alasca/.

A Bibliometric Analysis of Articles on Text Simplification: Sample of Scopus Database

Literacy is a term generally used for adults and young people. Basically, it is an acquisition that includes the process of reading, writing and understanding symbols in any language. While this concept, whose definition and scope has expanded over time, refers to people who can only say their names in the past, today it refers to individuals who can perform more functional skills. What is expected from today’s literacy, which is also referred to as functional literacy, is not just saying the name, but also understanding what you read and harmonizing these information with the environment. It is critical to create texts that are easier to interpret, especially for poor readers and individuals learning foreign languages, in today’s world where reading comprehension and correct use of information have become extremely important. In this context, the text simplification method, which is one of the text modification methods, comes to the fore. In accordance with these information, this current study aims to presents the bibliometric analysis of articles on text simplification, which are published in journals indexed in Scopus database in the field of social sciences. The data set of the study consists of 194 articles on text simplification published in journals scanned in the field of social sciences and scopus database. These 194 articles were examined in terms of different variables. The research is generally a descriptive study and document analysis method was used as a method. In the data analysis stage of the research, VOSviewer visualization software version 1.6.16 was used. According to the results obtained from the study, the most articles on text simplification were written in 2020 (f: 21), most cited article is “Interpretation as Abduction” (f: 363) written by J.R Hobbs, M.E Stickel, D.E. Appelt and P. Martin. Findings obtained from the research were shared in the form of tables, graphs and figures. The most common keywords that preffered by authors is “Simplification” (f: 18). The most cited institution is “Artifical Intelligence Center” (f: 363). The most cited journal is “Artifical Intelligence” (f: 363). The most published country is United States of America (f: 30). The most cited country is United States of America (f: 863). All findings obtained from the research were shared as tables and figures in the findings section.

Plain language as a tool used in the adaptation of texts for the purpose of teaching Polish as a foreign language — research overview

The aim of the article is to collect and present studies into text comprehensibility conducted so far. The author examines studies dealing with plain language, Polish language teaching and pedagogy. The article opens with an overview of the history of research into text difficulty in the world featuring various concepts of automatic measurement of text difficulty level. Next the author presents studies devoted to comprehensibility of utterances in Poland, as well as studies concerning text difficulty with regard to language teaching. This is followed by a discussion of the principles of the text simplification principle according to the Wrocław plain language model and in the teaching of Polish.The article closes with a short overview of publications containing simplified authentic texts used in the teaching of Polish as a foreign language.