Add deidentification project #4
Conversation
There was a problem hiding this comment.
Consider https://go.privitar.com/rs/588-MYA-374/images/2021-07-Privitar-Bristows-Intro_to_Anonymisation.pdf in current knowledge. How would the work sit in the against the techniques and issues mentioned here?
Consider https://ico.org.uk/media/1061/anonymisation-code.pdf in current knowledge. How would the work sit alongside this one?
Consider https://arxiv.org/pdf/2104.06523.pdf in current knowledge. How would the work sit alongside this narrative?
Suggestion: Consider moving focus from delivery of tool to investigating which tools available would give the confidence required to use the outputted data and if non-exist what this tool would look like (stretch to developing the tool itself). My concern is that any tool developed may end up being too narrow in focus or specific to an individual task. There are a great number of tools out there and many companies offering this service so I suppose I'm looking for what makes this worthwhile (answer: it's free and demonstrable) but then will we make something good enough in the time available?
Adding this after a chat with @danjscho - hope it reads okay? Feel free to change, obviously