10

Are there any tools which allow to measure the range of vocabulary in the text? If you know about such, please, provide their names, links, etc.

My aim is to find more varied in terms of range of vocabulary reading, so I will be able to increase lexical variety of my own speaking.

I've opened separate question to find out if such kind of rating already exists for modern Ukrainian literature and translations. Any rating by the vocabulary richness for modern Ukrainian literature works (or authors)?

UPD: I've searched a bit for more suitable term and found that "Vocabulary richness" describes better what I'm asking about.

  • Do you mean a global size of Ukrainian vocabulary? Could you update the question to make it more detailed? – ntk Feb 8 '17 at 19:16
  • 1
    @NazarTokar, I think, the question author needs to estimate the number of unique words used within a specific text — or something like that (some characteristic of a specific text) — if I understand correctly. – Sasha Feb 8 '17 at 20:34
14

If I understand your question correctly, that characteristic (range of vocabulary) is often implicitly calculated when tagging a text corpus. So some tools/services working with text corpora may provide such information.

For example:

  • The Ukrainian language corpus of Laboratory of Computer Linguistics of Institute of Philology of Taras Shevchenko National University of Kyiv has the “Статистика текстів” (statistic of texts) subsection. It allows to see lexeme statistic of different works by expanding a category (▹) at the left, choosing a work and clicking “Частотний словник лексем” (frequency dictionary of lexemes):

    screen-shot of MOVA.info

    I suppose that “всього лексем” (number of lexemes) may be a value you need.

    Still, it can only show you statistics of already pre-analyzed texts, without possibility to "feed" the tool with an arbitrary text for getting its (text) statistics.

  • Some other services for Ukrainian corpora are listed on the online sources meta page. Still, I don't know whether they can work with an arbitrary text and whether they provide such information at all.

There are tools that allow you to tag an arbitrary text corpus yourself. For example, one of them is LanguageTool (e.g. tagging a corpus using LanguageTool; for example, Leipzig Corpora Collection uses LanguageTool as part-of-speech tagger). Still, doing that maybe uneasy for unexperienced user and, besides this, the value you need may not come immediately from manipulations with LanguageTool themselves.

P.S.: However, I may misunderstand you; please notify me in that case.

  • @y_vyshnevska, BTW, if you consider an answer as OK, you should accept it by clicking ✓ at the left of it. Or, if you consider that it has some drawbacks, it's better to write what info is missing. Or course, it's not an obligation, you can leave the answer unaccepted as well. Still, IMHO, it doesn't bring any significant benefit (to grab more attention for your question it's better to edit it (clarify) and maybe even propose bounty — just leaving it unaccepted isn't enough for it). BTW accepting isn't closing — users still can post answer after. – Sasha Feb 13 '17 at 20:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.