How can I extract the reason of the legal compensation from a court report?

I'm working on a project (court-related). At a certain point, I have to extract the reason of the legal compensation. For instance, let's take these sentences (from a court report)

Order mister X to pay EUR 5000 for compensation for unpaid wages

and

To cover damages, mister X must pay EUR 4000 to mister Y

I want to make an algorithm that is able from this sentence to extract the motive of legal compensation. For the first sentence

Order mister X to pay EUR 5000 for compensation for unpaid wages

the algorithm's output must be "compensation for unpaid wages" or "compensation unpaid wages ".

For the second sentence, the algorithm's output must be "cover damages". Output can be a string or a list of string, it doesn't matter.

As I'm not an NLP expert (but I have already worked on a project on sentiment analysis, so I know some stuff about NLP), and there are so many articles, I don't know where to start.

I'm working on French texts, but I can get away with working on English texts.

Topic stanford-nlp data deep-learning nlp

Category Data Science

1
answered at

Can you clarify how consistent the structure of these sentences will be?

  • If the court reports contain very consistent phrasing, you might be able to get away with using a rule-based approach (or perhaps even regular expressions).
  • But, if the sentences would require human-level language abilities to parse, you would probably need to invest a lot of energy into training a model to accomplish this, which would also require a large(ish) set of labelled training data.

That said, you may be able to apply parts of speech (POS) tagging to isolate the right class of word in each sentence. I'm sure other options also exist, but I know NLTK accomplishes this task.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.