# You are your Metadata: Identification and Obfuscation of Social Media Users using Metadata Information

<https://www.ucl.ac.uk/~ucfamus/papers/icwsm18.pdf>

- Host: [ucl.ac.uk](https://links.thingelstad.xyz/hosts/ucl.ac.uk/)
- Saved: July 12, 2018 (2018-07-12T11:42:42Z)
- Record: https://links.thingelstad.xyz/links/9e2120e5a367/
- JSON: https://links.thingelstad.xyz/links/9e2120e5a367.json

## What Jamie wrote

More research that shows how precise identification cane be on metadata alone.

> We demonstrate that through the application of a supervised learning algorithm, we are able to identify any user in a group of 10,000 with approximately 96.7% accuracy. Moreover, if we broaden the scope of our search and consider the 10 most likely candidates we increase the accuracy of the model to 99.22%. We also found that data obfuscation is hard and ineffective for this type of data: even after perturbing 60% of the training data, it is still possible to classify users with an accuracy higher than 95%.

I still don't think this message is widely understood. When people are told that _only_ metadata is kept on their activities, they assume some level of anonymity. You should assume none.

## Facts

- Kind: [Paper](https://links.thingelstad.xyz/kinds/paper/)

## Appears in

- [WT62: Weekly Thing #62 / Jul 14, 2018](https://weekly.thingelstad.com/archive/62/), Notable Links 📌

## Saved the same week

- [Custom Perspectives for OmniFocus 3 - Learn OmniFocus](https://links.thingelstad.xyz/links/e431e5ffc8de/) (learnomnifocus.com, July 12, 2018)
- [Perspective Icons for OmniFocus 3, by Josh Hughes](https://links.thingelstad.xyz/links/d800ad1be7f3/) (omnifocusicons.josh-hughes.com, July 12, 2018)
- [9 Tips To Better Networking (Even For Introverts) - Toggl Blog](https://links.thingelstad.xyz/links/1935201e58ab/) (toggl.wpengine.com, July 12, 2018)
- [Road Trip](https://links.thingelstad.xyz/links/74af141b776c/) (roadtrip.glitch.me, July 12, 2018)
- [Leadership Style Colors | Lara Hogan](https://links.thingelstad.xyz/links/abfea8640fde/) (larahogan.me, July 12, 2018)

## Elsewhere

- Wayback Machine: https://web.archive.org/web/*/https://www.ucl.ac.uk/~ucfamus/papers/icwsm18.pdf