I am analysing a textual corpus with 34 obituaries of a recently deceased writer from the point of view of Discourse Analysis. I think that a quantitative analysis of the lexicon used would be interesting. Specifically, I am looking for a way of detecting the specific lexis of these texts (verbs, nouns, adjectives, collocations...). In some articles I have seen that they study the specific lexicon of a politician by comparing it with that of another politician. Here the comparison would be with a general language corpus such as CREA (Corpus de Referencia del Español Actual). I would be grateful for ideas about methods (formulas, programmes...) to measure lexical specificity.
Hello. If I understood well, you are looking for a POS (Part-Of-Speech) Tag software. If this is the case, have a look at this website: https://www.laurenceanthony.net/software/tagant/ (read the Help file).
Good luck!
En Iramuteq puedes analizar frecuencias absolutas y relativas de verbos, adjetivos, sustantivos lematizados o no. El programa identifica las categorías gramaticales para el español, inglés, francés, entre otras. También hace análisis factorial de correspondencias, que es muy útil para esas comparaciones de similitudes o diferencias con corpus de referencia.
The other answers presented here look great, yet, as always, there are oftentimes many ways to tackle these problems. You could use software like Iramuteq, as Andrea Torres Perdigón noted, to lemmatize and/or tag by lexical category. If you feel a little bit more comfortable with Python, then you could also use NLTK (the more common and heavier package) or TextBlob (the less common but easier package) to do both of these. Absolute and relative lexical frequency, collocations, concordances, etc. are all easily accomplished in both of those packages, but you might prefer using something like AntConc, which will allow you to load each of the obituaries in separately, i.e. as separate files, so that the wires never "cross."
Thank you so much for all your ideas and suggestions. I will explore each one!
Hi,
Thanks alot for your question and sharing interest with me concerning discourse analysis. It is so important and fruitful field. As far as quantitative analysis is concerned, you can use (AntConc) corpus programme. It is so useful in analysing texts and save your effects and time. You can search about it in google and download it in your computer. Then you can enter any text you want to analyse it in terms of verbs, adjectives or whatever you like. With regards
There is an online programme called TAALES which can do this for you (amongst other lexis measurements).
Hello,
It would be interesting to use Lexico3 to identify the *main* key word of these obituaries. See If that key word is used in CREA.
Lexico3 as text analyser has the possibility to display that key word in its collocation, as well. This will allow you to generate/predict the most probable (and least probable) social uses of that key word in obituaries. Thus, the linguistic context of occurence of that key word will be enlightened. The most compatible categories (verb, noun, adjective, ...) of other words likely to be associated or used along with that key word will be identified through a careful observation of the text output.
Best regards
Nb.
The main key word will be the *content* word (not grammatical one) displaying the highest lexicometric frequency of use.
Its score will be generated automatically on the Lexico3 data chart/sheet right after text treatment and analysis.
- Badges
- Science topic
- Similar topics
- Cognitive Science
- Thinking