Compile information about the NLP de-id methods submitted to the 2014 i2b2 challenge

[tschaffter]

Motivation

• Could be the second NLP de-id method that we evaluate in the NLP Sandox POC.
• Why this method? It will be interesting to compare the performance of MCW method to the performance of the best-performing method in the 2014 i2b2 challenge.

Guidelines

• Create 2-3 slides that gather information relative to the best-performing method in 2014 i2b2 challenge (GH repo URL, description, contact information, etc.)
• Find the source code or binary of the method
• Run the method against the 2014 i2b2 challenge data
• Apply the evaluation script provided by the challenge organizer to evaluate the performance of the method (if same source code/binary than used during the challenge, the performance that we obtain should match the performance achieved by the method as reported by the challenge organizers).

Progress

• [x] Review 1st best performing method
• [x] Review 2ndbest performing method
• [x] Review 3rd best performing method

Next task

The following tasks will be tracked in new tickets:

• Dockerize this method so that we can evaluate it in the NLP Sandbox POC.

[tschaffter]

@yy6linda What we need is a spreadsheet that lists the methods submitted to the 2014 i2b2 NLP de-id challenge with the following information.

• Name of the method
• URL to the source code of the the method
• Email of one of the authors
• Performance of the method in the i2b2 challenge

The information included in this spreadsheet will be very useful to invite these authors when we launch the NLP DREAM Challenge.

[tschaffter]

@yy6linda uploaded the review of the best performaing method (Hui Yang et al. (2015) method) to the shared Google drive: https://docs.google.com/presentation/d/1DyUefegslXtK9RfRScyCcKy0UkwLaLrp

[tschaffter]

Here is the Google folder that includes the three presentations from Yao.