dheeraj7596/ConWea

Code for the paper "Contextualized Weak Supervision for Text Classification"

43
/ 100
Emerging

This project helps data scientists and machine learning engineers classify text documents using minimal labeled data. You provide a dataset of text sentences and a small list of 'seed words' for each category you want to identify. The system then automatically generates a text classifier, even for fine-grained categories, by understanding the context of words.

No commits in the last 6 months.

Use this if you need to categorize a large volume of text but have very few examples of what each category looks like, and manually labeling data is too time-consuming or expensive.

Not ideal if you already have a large, fully labeled dataset for your text classification task, as this project focuses on weak supervision with minimal labels.

text-classification natural-language-processing weak-supervision data-science machine-learning
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

52

Forks

20

Language

Python

License

MIT

Last pushed

Mar 25, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/dheeraj7596/ConWea"

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