Timeline for How does text classification reduce manpower costs?
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 24, 2019 at 12:21 | vote | accept | Air Christmas | ||
| Oct 24, 2019 at 12:20 | comment | added | Air Christmas | I believe that point 2 actually answers my question the most. I have also come across a method that uses the predictions of multiple models. If those models all produce the same prediction, then it is likely that one can treat that prediction as 'correct' and thus save cost. | |
| Oct 23, 2019 at 23:48 | comment | added | John Doucette | @nbro FWIW, most spam classification literature uses AUC as a measurement. They typically have 99.99% AUC. This means that they can obtain extremely high accuracy on both classes, despite the class imbalance | |
| Oct 23, 2019 at 19:57 | comment | added | nbro | @NeilSlater If the majority of the e-mail I receive is spam, then I do not expect two e-mails to be marked (on the same day) as spam when they are not, if the predictive system is really understanding something about the difference between spam and not spam. In other words, if 99% of the e-mails are spam, then you can just predict always spam and you'll get a nice accuracy. Clearly, this accuracy is highly misleading. | |
| Oct 23, 2019 at 19:32 | comment | added | Neil Slater | @nbro: You also need to account for the fact that the majority of the email you receive is likely to be spam. The spam that you correctly don't see counts for most of that 99.9% accuracy rating. | |
| Oct 23, 2019 at 17:26 | comment | added | John Doucette | @nbro You are correct, I had misremembered the statistics. It's been at 99.9% or higher for quite a long time (more than a decade). Even though the classifiers keep improving, spam authors keep using more sophisticated techniques, so it has the nature of an arms race. | |
| Oct 23, 2019 at 17:25 | history | edited | John Doucette | CC BY-SA 4.0 |
deleted 3 characters in body
|
| Oct 23, 2019 at 16:08 | comment | added | nbro | I don't think that the spam classification system associated with my mail client is so accurate. For example, yesterday, it's labeled two e-mails as junk, while it was not the case. | |
| Oct 23, 2019 at 14:46 | history | answered | John Doucette | CC BY-SA 4.0 |