Skip to content
About
Main content
Model

Why removing names does not remove identity

Identity was never in the identifier column. It was spread across the whole row.

Cardinality
possible values of attribute i
Population
people in the dataset

Recurring principle: Independence is assumed

How it behaves

One input on the horizontal axis, the others held fixed.

Data table
Share of people alone in their combination (584 million combinations), by population
Population NUnique on these attributes
100K100%
115K100%
133K100%
154K100%
178K100%
205K100%
237K100%
274K100%
316K99.9%
365K99.9%
422K99.9%
487K99.9%
562K99.9%
649K99.9%
750K99.9%
866K99.9%
1M99.8%
1.15M99.8%
1.33M99.8%
1.54M99.7%
1.78M99.7%
2.05M99.6%
2.37M99.6%
2.74M99.5%
3.16M99.5%
3.65M99.4%
4.22M99.3%
4.87M99.2%
5.62M99%
6.49M98.9%
7.5M98.7%
8.66M98.5%
10M98.3%
11.5M98%
13.3M97.7%
15.4M97.4%
17.8M97%
20.5M96.5%
23.7M96%
27.4M95.4%
31.6M94.7%
36.5M93.9%
42.2M93%
48.7M92%
56.2M90.8%
64.9M89.5%
75M87.9%
86.6M86.2%
100M84.3%
115M82.1%
133M79.6%
154M76.8%
178M73.7%
205M70.4%
237M66.6%
274M62.6%
316M58.2%
365M53.5%
422M48.6%
487M43.4%
562M38.2%
649M32.9%
750M27.7%
866M22.7%
1B18%
1.15B13.8%
1.33B10.2%
1.54B7.16%
1.78B4.76%
2.05B2.97%
2.37B1.72%
2.74B0.92%
3.16B0.445%
3.65B0.192%
4.22B0.073%
4.87B0.024%
5.62B0.007%
6.49B0.001%
7.5B0%
8.66B0%
10B0%

Worked examples, recomputed

Every number the mechanisms quote for this relationship is recomputed from it, so none of them disagrees with it.

Where it appears

Each mechanism states it in its own names. The general symbol each name stands for is shown with it.