MT webMeetup #7 Translation Risk Prediction

It's been a while since our last Meetup. All the more we are happy to announce that we have teamed up with NLP Zurich (https://www.meetup.com/de-DE/NLP-Zurich) and will be holding a webMeetup on machine translation.

Adam Bittlingmayer (CEO of ModelFront) will talk about how to build risk prediction systems for multilingual machine translation. We are delighted to have Martin Volk, professor of Computational Linguistics at the University of Zurich and CEO at TextShuttle, introducing the topic.

Please register (below) for your personal webMeetup link. We are looking forward to welcoming you!

👉 Please register to receive your webMeetup link:
https://us02web.zoom.us/webinar/register/WN_492MmvDDTFSKG7nPa-8KWQ

Agenda

17:50 Join the webinar
18:00 Martin Volk (University of Zurich): Introduction
18:20 Adam Bittlingmayer (ModelFront): Translation Risk Prediction
19:00 Q&A
19:20 Virtual Hugs and Kisses ⊂(◉‿◉)つ

***

Talk summary:
Catching bad translations is at the core of the machine translation research tasks of parallel corpus filtering, quality estimation and quality evaluation. What are the applications, challenges, approaches? Can machines catch bad human translations? And how is it inspired by or useful to other NLP problems?
Adam will answer questions about building a robust, scalable and massively multilingual system for translation risk prediction.

About the speaker:
Adam is CEO and co-founder at ModelFront. He is a language learner and software engineer with experience at startups, Google Translate, Google Play and Adobe. Beyond machine translation, his interests include input correction, language identification, transliteration and synthetic data generation.
https://www.linkedin.com/in/bittlingmayer/
https://bittlingmayer.org/

About ModelFront:
ModelFront is the world's first translation risk prediction API, supports 100+ languages and custom models and includes a console for easily uploading files for risk prediction with no code.
ModelFront is sharing hundreds of millions of risk predictions for open datasets and provides free credit for 1.1M characters to every new account.
https://modelfront.com/