3 ms·
They use sampling to "preserve a fixed ratio of positive to negative examples." I wonder what the ratio is... isn't the technique w/ class imbalances to up-samp
by eggie5 9y ago
They use sampling to "preserve a fixed ratio of positive to negative examples." I wonder what the ratio is... isn't the technique w/ class imbalances to up-sample/subsample to get a 50/50 ratio so your classifier generalizes?
On the topic of feature engineering: I think the use of those inline-emojis on a message are a good indicator of importance. That could be another feature for the logistic regression.
Overall seems like a fun modeling challenge. It's nice to see people not just throw deep learning at the problem...
- saghm 9y ago> I think the use of those inline-emojis on a message are a good indicator of importance. That could be another feature for the logistic regression. Upon first reading that, I thought you meant that more emojis implies more importance, and I was very curious about what kind of conversations you and your peers were having on Slack