11 ms·

You have to be more careful than that. You need to write some script to follow random links on HN while you're away from your computer, so when you're off doin

by jfoutz 17y ago

You have to be more careful than that. You need to write some script to follow random links on HN while you're away from your computer, so when you're off doing risky things you won't change your profile. You need to occasionally forget your cell phone, so when you're doing risky things, forgetting your cell phone won't change your profile. If you have a car, you need to pass through the intersections with red light cameras at the normal times, but no extra times, so you don't change your profile. Does visa buy dns lookup histories? or cellphone active tower histories? or car intersection histories? probably not. yet. better to build the habits now though.

boredguy8 17y ago
Even better would be to have a probabilistic weight to these various items. Computers are way better at finding patterns and irregularities in patterns than we are at making irregularities. If I were to "accidentally" forget my phone based on my own decision, it would probably cluster in some way, say, forgetting it every 17 days +/- 1, or on Thursdays more than any other day, or something. So create a profile, then use RNG to match yourself to that profile.
4ensic 17y ago
Stepwise linear regression is how they do it. All statistical packages marketed since the late 1970's include the procedures. It's what the AI craze of the early 1990's used in their black box apps. Basically you associate a narrow set of outcomes (Filing for divorce, filing for bankruptcy, reporting a stolen credit card) and a mountain of possible predictors (store type, store name, store location, number of purchases from store in the billing cycle, ranges of purchases from a given store within the cycle). The stat program builds a regression equation out of the raw data by literally predicting trying to predict the past from data in the further past. The cool thing is that you can't put in too many variables (as in you don't have to know if something may be related or not) since the uncorrelated variables will simply vanish in the final equation.

by jfoutz 17y ago

boredguy8 17y ago

Even better would be to have a probabilistic weight to these various items. Computers are way better at finding patterns and irregularities in patterns than we are at making irregularities. If I were to "accidentally" forget my phone based on my own decision, it would probably cluster in some way, say, forgetting it every 17 days +/- 1, or on Thursdays more than any other day, or something. So create a profile, then use RNG to match yourself to that profile.

4ensic 17y ago

Stepwise linear regression is how they do it. All statistical packages marketed since the late 1970's include the procedures. It's what the AI craze of the early 1990's used in their black box apps. Basically you associate a narrow set of outcomes (Filing for divorce, filing for bankruptcy, reporting a stolen credit card) and a mountain of possible predictors (store type, store name, store location, number of purchases from store in the billing cycle, ranges of purchases from a given store within the cycle). The stat program builds a regression equation out of the raw data by literally predicting trying to predict the past from data in the further past. The cool thing is that you can't put in too many variables (as in you don't have to know if something may be related or not) since the uncorrelated variables will simply vanish in the final equation.