Using Machine translation, can I obtain a very compressed version of a sentence, eg. I would really like to have a delicious tasty cup of coffee would be translated to I want coffee Does any of the NLP engines provide such a functionality?
I got a few research papers that does paraphase generation and sentence compression. But is there any library which has already implemented this?
In producing compressions, you are free to delete words, add new words, substitute them or reorder them. While doing the task you will find that word deletion is the most frequent compression operation. You should use substitution, insertion, and reordering operations only if they render the original sentence shorter.
Sentence compression is the task of compressing a long sentence into a short one by deleting redundant words. Paper.
If your intention is to make your sentences brief without losing important idea from that sentences then you can do that by just extracting triplet subject-predicate-object.
Talking about tools/engine, I recommend you to use Stanford NLP. Its dependency parser output already provides subject and object(if any). But you still need to do some tuning to get desired result.
You can download Stanford NLP and learn sample usage here
I found paper related to your question. Have a look at Text Simplification using Typed Dependencies: A Comparison of the Robustness of Different Generation Strategie
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With