1
$\begingroup$

I'd like to build something (ideally in Python) that can translate an English sentence into "basic" English.

Are there any free/open-source tools/frameworks that can help? If not, what kind of steps can help solve this problem? (e.g., building on existing work like WordNet or pre-trained word embeddings).

By "basic" I mean things like being concise (avoiding unnecessary verbiage) and using well-known words (without compromising too much on meaning). I might even consider "broken" English, where verbs for example are lemmatised to their root.

The idea is to get something akin to what Wiki have achieved with their basic English versions of articles. And to give one more example, simplifying sentences is one of the services provided by the Grammarly app.


For example:

English:

I hurriedly made my way over to the store in order to purchase various garments from there

"Basic" English

I quickly went to the shop to buy clothes

"Broken" English

I quick go shop buy clothes

$\endgroup$
3
  • 1
    $\begingroup$ Does this look like text summarization? There could be pretrained models for that. $\endgroup$ Commented May 26, 2021 at 17:30
  • $\begingroup$ This task can also be seen as a special type of machine translation (from "High English" to "Everyday English") so any machine translation tool (with the appropriate language models given) can do the task $\endgroup$
    – Nikos M.
    Commented May 27, 2021 at 14:09
  • $\begingroup$ After a bit of research on "text summarization" and all the rabbit holes therein, I may have a plan... given an uncommon word in the sentence (e.g., "stick it in the oven"), BERT may allow me to find a list of context-based substitutions ("put", "place", "bung", etc.); from there it may be a case of picking the most commonly occurring word using known word frequencies (e.g., "put") and then making a straight substitution? $\endgroup$
    – Ben
    Commented May 27, 2021 at 15:57

1 Answer 1

1
$\begingroup$

Text simplification is not as popular as other generative tasks (e.g. summarization, translation). Nevertheless, you may try the approach of the Keep It Simple: Unsupervised Simplification of Multi-Paragraph Text paper (ACL'21). The authors released their source code and their models.

$\endgroup$

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