CWID-hi: A Dataset for Complex Word Identification in Hindi Text

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 5627–5636 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 5627 CWID-hi: A Dataset for Complex Word Identification in Hindi Text Gayatri Venugopal♢, Dhanya Pramod∗, Ravi Shekhar† ♢Symbiosis Institute of Computer Studies and Research, Symbiosis International (Deemed University), Pune, India ∗Symbiosis Centre for Information Technology, Symbiosis International (Deemed University), Pune, India †Cognitive Science Research Group, Queen Mary University of London, London, UK gayatri.venugopal@sicsr.ac.in, dhanya@scit.edu, r.shekhar@qmul.ac.uk Abstract Text simplification is a method for improving the accessibility of text by converting complex sentences into simple sentences. Multiple studies have been done to create datasets for text simplification. However, most of these datasets focus on high-resource languages only. In this work, we proposed a complex word dataset for Hindi, a language largely ignored in text simplification literature. We used various Hindi knowledge annotators for annotation to capture the annotator’s language knowledge. Our analysis shows a significant difference between native and non-native annotators’ perception of word complexity. We also built an automatic complex word classifier using a soft voting approach based on the predictions from tree-based ensemble classifiers. These models behave differently for annotations made by different categories of users, such as native and non-native speakers. Our dataset and analysis will help simplify Hindi text depending on the user’s language understanding. The dataset is available at https://zenodo.org/record/5229160. Keywords: lexical simplification, complex word identification, dataset, classification, Hindi 1. Introduction The significance of accessible technology has in- creased manifold today for creating an inclusive en- vironment. Accessibility not only helps improve the quality of life of persons with impairments but also benefits individuals without impairments (Henry, 2006). We can associate accessibility with physical devices as well as with the content we consume in our day-to-day lives. We can achieve accessibility of textual content by simplifying it. Text simplification refers to the process of modifying text in such a way that it becomes easier for the reader to comprehend it. It is a sub-field of natural language processing, that has proven to be useful to children (De Belder and Moens, 2010), readers with language impairments (Ca- plan, 1992; Carroll et al., 1999; Carroll et al., 1998), readers with low literacy levels (Candido Jr et al., 2009), and non-native speakers (Paetzold and Specia, 2016c). Conceptual simplification, elaborative mod- ification, syntactic simplification, and lexical simpli- fication are different ways by which we can simplify text (Siddharthan, 2014). In this work, we focus on lexical simplification of Hindi text, i.e., the process of identifying complex words in a given text and substituting them with their simpler synonyms based on the context of the target complex word (Paetzold and Specia, 2017). The term ’complex’ in this context does not indicate syntactic or morpho- logical complexity. A complex word is defined as a word that is difficult to understand by the reader (Finni- more et al., 2019; Yimam et al., 2017a). Shardlow (2014) has demonstrated the steps in a lexical simplifi- cation pipeline, which is …

Content truncated to 500 words