Inspiration

What it does

How I built it

Challenges I ran into

Prototype and gradually improve the annotation process to get high-quality human annotation quickly.

Domain-specific and data-scarce text classification problem. Many out-of-vocabulary words due to the hardness and cost to get enough useful and accessible in-domain data.

Develop and improve transfer learning techniques to enhance classifiers.

Accomplishments that I'm proud of

What I learned

What's next for IBM-Medical Abstract Annotation

Built With

Share this project:

Updates