Hi

I am seeking for a way to extract linguistic rules out of given ground truth sentences, then apply these rules to new sentences so as to classify them.

For example, considering "You should add a new button" as a sentence, the following rule can be extracted:

• “[someone] should add [something]”

More examples would be:

• "[something] would be a good idea/thing"
• "[something] seems to be an issue"

So is there any tool that could help me with that or at least a systematic solution for mining linguistic rules?

Please let me know if you had any idea.

Thanks a lot.