Unscrambling messages using NLP

Web: https://www.reddit.com/r/deeplearning/comments/sb3noz/unscrambling_messages_using_nlp/

I am currently working on an operation that requires unscrambling messages.

For example, consider the sentence:

I like coding in Python and R (original)

It now reads:

like I python coding in R and (scrambled)

I need to find a way of unscrambling these messages.

I thought about tokenising and tagging the words (e.g. adjective, noun etc.) in each message and then somehow generating a syntactically correct sentence using this information.

Does anyone know of an NLP library in Python …

deeplearning messages nlp