|
|
|
|
|
|
by Karrot_Kream
1299 days ago
|
|
|
When RSS was hot stuff circa 2005, a friend on IRC and I were working on a bayesian classifier for RSS entries and it worked just like this. +1 trained it as "ham" and -1 trained it as "spam". It failed (among other reasons) because we were both really young and didn't have enough mathematial/computational maturity. With modern ML techniques, this doesn't seem hard to do at all. But you'd need access to the actual data stream which is something that the big social networks don't seem willing to provide. |
|
|