4 ms·
It seems to be some sort of common principal component analysis. E.g. A likes a. B likes b. B likes c. A maybe likes c. But with many more nodes. I think the
by rightbyte 3y ago
It seems to be some sort of common principal component analysis.
E.g.
A likes a.
B likes b.
B likes c.
A maybe likes c.
But with many more nodes. I think they use it for recommendations and subgenre sorting etc.