T

text-processing

Converts DOCX documents into a structured JSON representation

A service for extracting action parameters occurring in phrases/sentences within the processed document

The service is designed to determine the grammatical properties of an action, such as tense, mood, etc.

The service segments natural language text into sentences and then extracts individual phrases built around identified activities.

Text Vectorization (as pypi package) divides text into tokens and then saves individual tokens in vector form

spaCy pipeline for Polish legal texts with a Morfeusz 2 and Toygger tokenizer, legal sentence boundaries and a legal NER model