ghxm / eucy Goto Github PK

• could provide easy access to all euCy metrics
• could provide reference to used nlp objects which can then be used for functions that require re-running (such as modify)

Implement fuzzy matching of entities using spacy patterns and https://github.com/gandersen101/spaczz

see e.g. #9 but also others.

Options are to either

• try to parse outer structure and account for amending acts by setting specific flag or similar in annotation
• parse inner structure -> e.g. inserted article within article counts as extra article (messy)

• euplexer version:
• Python version:
• Operating System:

Description

Describe what you were trying to get done.
Tell us what happened, what went wrong, and what you expected to happen.

What I Did

Paste the command(s) you ran and the output.
If there was a crash, please include the traceback here.

pip
poetry
easy_install

https://github.com/Lexparency/eurlex2lexparency

Add argument option to classes and function, default to euplexre.

E.g. https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=cellar:f9e18436-03af-449a-80aa-99854c6c7615

Proposals, Final texts
Time periods
Non/amending
Regulations, Directives, Decisions

Doesn't seep to recognized in https://github.com/ghxm/euCy/issues/9 but might be down to reports parsing not being implemented properly.

• euCy version:
• Python version:
• Operating System:

Description

Recital 2 not matched correctly (seems like text of an article is picked up.

What I Did

Proposal 2018_232.txt

Possible solution

Make sure individual elements are more or less adjacent

It shouldn't be.

remove import en_core_web_sm from eucy/eucy.py

• Blackstone sentencizer?
• What about footnotes? ('Having regard to the proposal from the Commission [4],')
• Convert _get_sentences function with min length etc to custom Sentencizer to use and adjust/replace over time?

Objective

Individual parts of laws (e.g. spans in the .spans SpanGroups should be editable.

Aims

• Obtain a spaCy document with updated text while keeping all annotations/elements

Solutions

A. Markup text and export marked text

• Store replacement text in ._.replacement_text attribute
• Import into spaCy/euCy with special markup reader for element detection

Problem: When re-reading into spaCy, markup will become part of document

B. Export text and annotation separately

• Store replacement text in ._.replacement_text attribute
• Export annotation in standard spaCy JSON

Problems:

• How to export the text and
• update the spans (when a previous span changes, start of next one also changes)

C. Split text and recreate doc at every (changes) span to obtain text + annotation (can then continue with e.g. 2.)

→ https://stackoverflow.com/a/75300856/5565500

Works at token level?

Problem: Problem might be computationally heavy

https://onlinelibrary.wiley.com/doi/epdf/10.1111/j.1540-5907.2009.00383.x

If replacement does not have (1) but replaced part has (1) ...