21–30 of 99 posts
If you're interested in tools such as Amnesia, you might also want to take a look at ARX, which supports much more anonymization methods, including Differential Privacy: https://arx.deidentifier.org https://github.com/arx-deidentifier/arx Disclosure: I'm the main author of ARX.
This is not easy. This is not an easy thing to do and protect the data. Please don’t call this easy.
Earlier quoted context omitted.
Then get informed consent
That's a must. In some countries this is now automated at the government level, Belgium for instance has an excellent consent mechanism for medical data.
I'd also point out that there is a huge amount of COVID data being shared at the moment and, outside of vaccine trials, I'd be pretty certain a lot of it is not under any sort of informed consent. (As is true of a lot of population statistics generally.)
> Amnesia Desktop Version 1.2.2
> Available for Windows and Linux
> [ mockup that is supposed to resemble a macOS app ]
:-P
If you're interested in tools such as Amnesia, you might also want to take a look at ARX, which supports much more anonymization methods, including Differential Privacy: https://arx.deidentifier.org https://github.com/arx-deidentifier/arx Disclosure: I'm the main author of ARX.
And has a huge advantage: it runs local.
Fair warning: anonymization is a hard problem. It is never easy, and you'd be surprised how many bits can leak out of what you thought was properly anonymized data. If you are using data for test purposes please use generated data, not anonymized data. This has the additional advantage that there is no potential path for live data to end up on a developers machine. added in edit: And also realize that just using a se…
What about building machine learning models that make predictions on said data? Can't just test on fake data.
Fair warning: anonymization is a hard problem. It is never easy, and you'd be surprised how many bits can leak out of what you thought was properly anonymized data. If you are using data for test purposes please use generated data, not anonymized data. This has the additional advantage that there is no potential path for live data to end up on a developers machine. added in edit: And also realize that just using a se…
Earlier quoted context omitted.
And has a huge advantage: it runs local.
Amnesia doesn't...? Then what's the download link and github page for?
In my opinion, this source code was put out to appear "open" to the H2020 Programme, but they have no intention in actually helping users run the code locally.
Fair warning: anonymization is a hard problem. It is never easy, and you'd be surprised how many bits can leak out of what you thought was properly anonymized data. If you are using data for test purposes please use generated data, not anonymized data. This has the additional advantage that there is no potential path for live data to end up on a developers machine. added in edit: And also realize that just using a se…
Agree with your point about anonymization, but here is a call out for generated data too: if the objective is to build a model, you might end up losing information, or worse, your model might end up modelling the assumptions in the generation process.