Neural Pattern Model
The AI layer that turns a long list of results into four simple groups.
In one sentence
The neural pattern model measures every number's past behaviour and places it into one of four groups.
What it measures
- Frequency: how many times a number has been drawn in the stored record.
- Recency: how many draws have passed since it last appeared.
Combining the two gives each number a profile. Numbers with similar profiles end up in the same group.
What it doesn't do
It doesn't know what will be drawn next, because no one does. A number that has been quiet is not "due". Read more in our guide to the number patterns people pick.
Where its work shows up
The group ring in the chart suite is a direct picture of the model's four groups. The recommendation generator also uses the groups when it builds a set.
Doing this grouping by hand would take hours. Apollo AI finishes the sorting before you even look at the screen.