6 ms·
> I even know the percentage roughly. No, you don't. This "principle" is not a law of physics, it's just a rule of thumb that may or may not be wildly off the
by throwamon 4y ago
> I even know the percentage roughly.
No, you don't. This "principle" is not a law of physics, it's just a rule of thumb that may or may not be wildly off the mark for an arbitrary distribution. Its common usage is like statistics horoscope.
- andreilys 4y agoHaving worked as a data scientist at multiple companies (From FANG to startup), the first thing I look at when I get my hands on data is the existence of the Pareto principle. I still haven’t found one company where this principle didn’t show up.
- tomp 4y agoSo you massage your features until they can produce a 20-80% split? Very scientific.
- throwamon 4y agoWhat does this prove? If you have lots of data and dimensions, I bet you could just as likely find distributions that are roughly 50/50, 60/40, 90/10, 100/0 if you looked for them.
- dado3212 4y agoAgreed. It’s not necessarily 80/20, it’s just that power laws show up a LOT. 90/10, 99/1, etc.
- hbarka 4y agoThe 50/50 principle always shows up in my histogram with interval of two.