Library devoted to Document level Attitude and Relation Extraction for text objects with entity-linking (EL) API support
Project description
AREkit 0.22.1
AREkit (Attitude and Relation Extraction Toolkit) -- is a python toolkit, devoted to document level Attitude and Relation Extraction between text objects from mass-media news.
Description
This toolkit aims to solve data preparation problems in Relation Extraction related taks, considiering such factors as:
- 🔗 EL (entity-linking) API support for objects,
- ➰ avoidance of cyclic connections,
- :straight_ruler: distance consideration between relation participants (in
terms
orsentences
), - 📑 relations annotations and filtering rules,
- *️⃣ entities formatting or masking, and more.
Using AREkit you may focus on preparation and experiments with your ML-models by shift all the data-preparation part onto toolset of this project (tutorial). In order to do so, we provide:
- :file_folder: API for external collection binding (native support of BRAT-based exported annotations)
- ➿ pipelines and iterators for handling large-scale collections serialization without out-of-memory issues.
- evaluators which allows you to assess your trained model.
AREkit complements the OpenNRE functionality since document-level RE setting is not widely explored (2.4 [paper]). The core functionality includes (1) API for document presentation with EL (Entity Linking, i.e. Object Synonymy) support for sentence level relations preparation (dubbed as contexts) (2) API for contexts extraction (3) relations transferring from sentence-level onto document-level, and more. It providers contrib modules of neural networks (like OpenNRE) applicable for sentiment attitude extraction task.
Installation
pip install git+https://github.com/nicolay-r/AREkit.git@0.22.1-rc
Download Resources
from arekit.data import download_data
download_data()
Applications
- ARElight [site] [github]
- Infer attitudes from large Mass-media documents or sample texts for your Machine Learning models applications
Papers
- Frame-Based attitude extraction workflow for news processing [code]
- Represents an attitude annotation workflow based on RuSentiFrames lexicon which is utilized for news processing;
- Neural Networks Applications in Sentiment Attitude Extraction [code]
- Neural Networks application for attitude extraction from analytical articles;
- BERT-based model utils for Sentiment Attitude Extraction task [code]
- Analytical news formatter for BERT-based models;
Related Frameworks
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.