From: Where you go is who you are: a study on machine learning based semantic privacy attacks
Check-in data | POI data | Method | Place categorization | User profiling | ||
---|---|---|---|---|---|---|
Accuracy | Profiling error | Top-5 identifi-cation accuracy | Privacy loss (median) | |||
Foursquare (NYC and Tokyo) | Foursquare | Random | 0.159 | 0.235 | 0.003 | 1.000 |
 |  | XGB (temporal) | 0.344 | 0.205 | 0.023 | 1.112 |
 |  | spatial join | 0.459 | 0.208 | 0.183 | 2.296 |
 |  | XGB (spatial) | 0.580 | 0.140 | 0.328 | 5.574 |
 |  | XGB (spatio-temporal) | 0.616 | 0.124 | 0.404 | 11.001 |
 | OSM | Random | 0.157 | 0.235 | 0.002 | 1.000 |
 |  | Spatial join | 0.224 | 0.341 | 0.013 | 1.051 |
 |  | XGB (temporal) | 0.328 | 0.206 | 0.022 | 1.116 |
 |  | XGB (spatial) | 0.476 | 0.172 | 0.117 | 1.949 |
 |  | XGB (spatio-temporal) | 0.503 | 0.157 | 0.160 | 2.316 |
Yumuv study (Switzerland) | Foursquare | Random | 0.370 | 0.367 | 0.010 | 1.000 |
 |  | Spatial join | 0.453 | 0.698 | 0.009 | 1.000 |
 |  | XGB (spatial) | 0.541 | 0.331 | 0.034 | 1.095 |
 |  | XGB (temporal) | 0.553 | 0.333 | 0.038 | 1.108 |
 |  | XGB (spatio-temporal) | 0.618 | 0.293 | 0.065 | 1.225 |
 | OSM | Random | 0.372 | 0.367 | 0.010 | 1.000 |
 |  | Spatial join | 0.384 | 0.698 | 0.009 | 1.000 |
 |  | XGB (spatial) | 0.541 | 0.332 | 0.034 | 1.093 |
 |  | XGB (temporal) | 0.553 | 0.333 | 0.038 | 1.108 |
 |  | XGB (spatio-temporal) | 0.623 | 0.289 | 0.060 | 1.256 |