What software/tools/algorithm can look at 'daily/weekly/monthly real world data' to calculate a "numeric score". Says, five is 'normal'. Four or six is 'slightly up or down' from established trend. One and nine is really exceptional.
Real world data has general trend/pattern plus random noise. (weather) Temperature, electricity demand, number of train rider at every hour of a day, ....
Some old days AI used rules-based system. How is the latest method?
I saw a youtube video. People feed hundreds of photo of a box from different angle/zoom and part of the box, into a Google trainer and it can achieve quite good accuracy after 'seeing' hundreds of photo.
Google says, "A rule-based system in AI is a system that applies human-made rules to store, sort and manipulate data. In doing so, it mimics human intelligence. Rule based system in AI require a set of facts or source of data, and a set of rules for manipulating that data."
Many thanks
Real world data has general trend/pattern plus random noise. (weather) Temperature, electricity demand, number of train rider at every hour of a day, ....
Some old days AI used rules-based system. How is the latest method?
I saw a youtube video. People feed hundreds of photo of a box from different angle/zoom and part of the box, into a Google trainer and it can achieve quite good accuracy after 'seeing' hundreds of photo.
Google says, "A rule-based system in AI is a system that applies human-made rules to store, sort and manipulate data. In doing so, it mimics human intelligence. Rule based system in AI require a set of facts or source of data, and a set of rules for manipulating that data."
Many thanks
Statistics: Posted by David16962 — Sun Sep 08, 2024 7:18 am