thunlp/OpenNRE
OpenNRE
OpenNRE is a natural language processing library and neural relation extraction framework designed to transform unstructured text into structured relational data. It serves as a toolkit for identifying relationship types between entities and generating entity-relation-entity triples to populate and expand knowledge bases.
The framework provides tools for both supervised and distantly supervised relation extraction, allowing neural models to be trained on labeled datasets or via automated pipelines that align knowledge base triples with raw text.
The project covers a full information extraction pipeline, including transformer-based text encoding, relation inference, and the output of structured triples for knowledge graph construction.
Features
- Entity and Relation Extraction - Identifies relational facts between entities in plain text to produce structured knowledge triples.
- Supervised Relation Extraction - Implements a system for training neural models to recognize specific entity relationships using labeled or distantly supervised datasets.
- Relation Extraction Training - Trains neural networks to identify and extract specific relationships between entities using various supervision methods.
- Relation Inference Engines - Predicts specific relationship types and confidence scores for given sentences using pre-trained models.
- Natural Language Processing Libraries - Provides a set of tools for analyzing human language to transform unstructured text into structured relational data.
- Supervised Learning Frameworks - Provides a framework for training neural networks on manually labeled datasets to map entity pairs to relationship categories.
- Supervised Text Learning - Trains neural network models to recognize specific relationship types using labeled datasets of entity pairs.
- Information Extraction - Implements a full information extraction pipeline that converts natural language into machine-readable relational facts.
- Transformer Text Encoders - Uses transformer-based encoders to convert raw text into dense vector representations for relation extraction.
- Modular Architectures - Provides a modular neural architecture allowing the use of interchangeable pre-trained language model encoders.
- Distant Supervision Pipelines - Implements automated pipelines to label training data by aligning knowledge base triples with raw text.
- Relation Scoring Mechanisms - Calculates the probability of a relationship existing between two identified entities based on contextual embeddings.
- Knowledge Graph Construction Tools - Turns raw text into a network of entities and relations to build structured knowledge graphs.
- Triple Formatters - Formats model predictions into subject-predicate-object sets for direct integration into knowledge graphs.
Star history
How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.
