pmichel31415/mtnt

Code for the collection and analysis of the MTNT dataset

33
/ 100
Emerging

This project helps machine translation researchers and developers create and analyze datasets for translating "noisy" real-world text, like social media posts. It takes raw text data (e.g., Reddit comments) and processes it to identify characteristics of noisy language, then prepares it for use in machine translation experiments. Researchers focusing on improving machine translation performance for less formal, user-generated content would use this.

No commits in the last 6 months.

Use this if you are a machine translation researcher needing to build or analyze datasets of informal, real-world text for training and evaluating translation models.

Not ideal if you are looking for a ready-to-use machine translation application or a tool for translating clean, formal documents.

machine-translation natural-language-processing data-collection text-analysis social-media-data
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

56

Forks

4

Language

Python

License

MIT

Category

scraper

Last pushed

Apr 02, 2019

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/perception/pmichel31415/mtnt"

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