A Thesaurus-based Sentiment Lexicon for Danish: The Danish Sentiment Lexicon

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 2826–2832 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 2826 A Thesaurus-Based Sentiment Lexicon for Danish: The Danish Sentiment Lexicon Sanni Nimb1, Sussi Olsen2, Bolette S. Pedersen2, Thomas Troelsgård1 The Society for Danish Language and Literature1, Centre for Language Technology, NorS, University of Copenhagen2 Christian Brygge 1, 1219 København K1, Emil Holms Kanal 2, 2300 Copenhagen S2 {sn,tt} @dsl.dk, {saolsen,bspedersen}@hum.ku.dk Abstract This paper describes how a newly published Danish sentiment lexicon with a high lexical coverage was compiled by use of lexicographic methods and based on the links between groups of words listed in semantic order in a thesaurus and the corresponding word sense descriptions in a comprehensive monolingual dictionary. The overall idea was to identify negative and positive sections in a thesaurus, extract the words from these sections and combine them with the dictionary information via the links. The annotation task of the dataset included several steps, and was based on the comparison of synonyms and near synonyms within a semantic field. In the cases where one of the words were included in the smaller Danish sentiment lexicon AFINN, its value there was used as inspiration and expanded to the synonyms when appropriate. In order to obtain a more practical lexicon with overall polarity values at lemma level, all the senses of the lemma were afterwards compared, taking into consideration dictionary information such as usage, style and frequency. The final lexicon contains 13,859 Danish polarity lemmas and includes morphological information. It is freely available at https://github.com/dsldk/danish-sentiment-lexicon (licence CC-BY-SA 4.0 International). Keywords: sentiment lexicon, linked data, thesaurus 1. Introduction The Danish Sentiment Lexicon was published in 2021, and is freely available at https://github.com/dsldk/danish- sentiment-lexicon (licence CC-BY-SA 4.0 International). The lexicon contains approx. 14,000 Danish polarity lemmas and includes morphological information.This paper describes how the lexicon was compiled by use of lexicographic methods and based on linked lexical data in line with other formal lexicons compiled in collaboration between The Society for Danish Language and Literature (DSL) and Centre for Language Technology at University of Copenhagen (CST) the last two decades. In this case the linked data constitute The Danish Thesaurus DDB (Den Danske Begrebsordbog, Nimb et al. 2014) linked at sense level to the comprehensive monolingual Danish dictionary, Den Danske Ordbog (DDO) and representing approx. 100,000 lemmas and more than 120,000 word senses. The linked dataset consisted of lemma senses listed in semantic order from the manually identified polarity sections in the thesaurus, combined with information from DDO on style, usage and frequency. In the same way as word senses have either neutral, positive or negative polarity, the titles of thesaurus sections convey a certain polarity. The idea was to identify the positive and negative thesaurus titles as the starting point for a sentiment lexicon for Danish with a much higher lexical coverage than existing Danish sentiment lexicons, and based not only on the subjective judgment of the word itself, but also …

Content truncated to 500 words