tv.um.es

The lexical list repository aims to provide an easy-to-find, easy-to-use one-stop-solution to
word lists. Corpus linguists, computational linguists and developers of technology-enhanced
language learning software frequently make use of various types of word lists. Textbooks for
learners of English also incorporate various lists with a list of irregular verb forms being
commonplace. The general service list and academic word list are two commonly-used yet
rather dated lists that feature regularly in academic articles. This may be due to ease of
access, however, rather than suitability of purpose.
Lists required for research or pedagogic purposes may be based on form, such as parts of
speech, particularly for closed sets, e.g. prepositions; or class of words, such as stative verbs,
e.g. love and live; or words semantically related, such as lexical sets, e.g. spicy, salty, bitter,
etc. However, to date, there is no single repository that provides free access to curated
categorized lists. This means that researchers need to create lists ad hoc.
The list creation process may involve searching for relevant lists offline in corpus grammars
and textbooks and online with multiple searches in the indexed web. Having spent many
hours searching for and compiling lists to help my own research agenda, and through
discussions with other researchers, it became clear that many researchers also create their
own lists. This project aims to fill this niche and meet the demand from researchers and
teachers by providing central shared repository.
This study details the creation of an open-access repository, currently housed on GitHub, that
users can access, download and contribute to lists for English. A wish-list of lists was created,
and the repository structured to enable researchers to commit and combine their lists. Lists
published in various corpus grammars and coursebooks were compiled and collated. These
lists from off-line sources were amalgamated with lists from online resources to create master
lists for each category. The creation process is documented for each word list to enable
verification of the word lists. Moderators monitor the quality of the lists and ensure that the
documentation is up to date. Lexical lists are systematically categorized and available in
multiple formats to reduce the technical entry barrier for those who want to conduct corpus
research but are less familiar with programming.
The current categories of lists include: part-of-speech lists: pronoun, preposition, etc.;
word class lists: attributive adjectives, predicative adjectives, postnominal adjectives,
gradable and limit adjective pairs, etc.; semantic category lists: cars, countries, cities; etc. and
lexical sets: kitchen, recipes and ingredients, etc. Each list provides details of the sources,
collection protocols, and any pertinent information. Some lists also contain regular
expressions for JavaScript and Python, which may be of use to some developers. This
repository actively seeks contributions from researchers worldwide, and so it is envisaged
that the quantity and quality of datasets will increase greatly over time as this resource gains
in popularity among the corpus and computational linguist communities.