Qznan/SpanKL

Code for paper: A Neural Span-Based Continual Named Entity Recognition Model

34
/ 100
Emerging

This project offers a system for identifying and categorizing specific terms, like names, organizations, or locations, within text documents as new types of terms emerge over time. It takes in text with various named entities and, through a continuous learning process, outputs text where these entities are accurately recognized and labeled. This is useful for computational linguists, NLP researchers, or anyone building information extraction systems that need to adapt to evolving terminologies.

No commits in the last 6 months.

Use this if you need to build or evaluate a named entity recognition (NER) system that can incrementally learn to identify new types of entities without forgetting previously learned ones.

Not ideal if you are looking for a ready-to-use, pre-trained NER model for a fixed set of entity types without any need for continual adaptation.

named-entity-recognition natural-language-processing information-extraction continual-learning text-analytics
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

18

Forks

3

Language

Python

License

Apache-2.0

Last pushed

Dec 11, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/Qznan/SpanKL"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.