Timeline for Machine learning - features engineering from date/time data
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 19, 2019 at 10:39 | comment | added | plpopk | Actuall as @PacoBarter said, one-hot encoding ignore the different distance between categories. This is not that easily tackle-able as these features are intrinsically phase info, while most machine learning models has no phase type input. Some DIY on distance metrics might do though. | |
| Mar 3, 2017 at 18:03 | comment | added | Paco Barter | Ok, I see. You're right, a continuos feature better captures the "proximity" quality of dates. My point is that there might be trends in the data for what the numeric values of dates are irrelevant (for example, a certain pattern of customer purchasing only in saturdays). Hence offering another point of view for dealing with dates. | |
| Mar 3, 2017 at 13:18 | comment | added | Sean Owen | In this instance, you effectively have columns like "is_12th" and "is_13th" which are, in the input space, unrelated, and unrelated to "is_1st", etc. As a continuous feature, it would correctly capture that the 12th and 13th are in some sense closer than 1st and 12th are. You are appealing to what a model might infer, but, I am talking about what the input features encode. | |
| Mar 2, 2017 at 23:33 | comment | added | Paco Barter | I can't see why it fails capturing the "proximity" of near dates. If you, for example, feed the binary vector to a NN it'll figure it out itself after proper training. Using binary vectors is only one way of representing categories. | |
| Mar 2, 2017 at 19:07 | comment | added | Sean Owen | This fails to capture relationships that probably exist, like, that the 14th and 15th of the month are 'similar'. To the extent that you believe that every day is literally different, you also believe that prediction about tomorrow is not possible. It's also not necessary to one-hot encode categoricals, not necessarily. | |
| Mar 2, 2017 at 18:53 | review | Late answers | |||
| Mar 3, 2017 at 1:42 | |||||
| Mar 2, 2017 at 18:38 | review | First posts | |||
| Mar 2, 2017 at 19:07 | |||||
| Mar 2, 2017 at 18:36 | history | answered | Paco Barter | CC BY-SA 3.0 |