Drop your questions for the Q&A event here

A Q&A event will be held on September 6th, 2023 from 1:30 - 2:30 PM ET via Zoom! The event will be recorded and shared with all solvers. If only a few people have questions, or if the submitted questions can all be answered simply, questions will be answered asynchronously via the challenge forum rather than via a live event.

You can register to attend the event using this form. If you do not have access to google forms, please send an email to info@drivendata.org.

Submit questions via the registration form or drop your questions here. Questions can focus on the subject area (injury prevention, older adult falls), the NEISS data, machine learning methods, or any other aspect of the challenge. All submitted questions will be answered either via the event or the forum.

Hi, :wave:
I have two questions:

  1. how the severity of fall is defined? Is it the ‘disposition’ or it’s about the diagnosis (+ body part)?
  2. Regarding provided OpenAI embeddings: just to confirm - Narratives have been lowercased before being passed through the model and nothing else was modified?
    Thank you!
1 Like

Thanks @mapl; we have the same question in mind:

How should we order these codes?

Our guess:
‘5 - HELD FOR OBSERVATION’: # mildly
‘6 - LEFT WITHOUT BEING SEEN’ # not severe at all

1 Like

Thanks for these questions! We’ll cover them at the event and will also follow up here on the forum with answers.

1 Like

There is no one correct way to operationalize severity, and part of the work of this challenge is finding effective methods of engineering features like those related to severity.

Related tip: In using disposition, the CDC team combines transfers to other hospitals with hospitalizations (given the low frequency of transfers).


This is correct. For the OpenAI embeddings on the data download page, the only preprocessing done was lowercasing.

1 Like

Super helpful. Thank you, Hannah!

Thank you, Hannah! :slightly_smiling_face: