Timeline for Predict when a user logins next
Current License: CC BY-SA 4.0
20 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 6, 2019 at 17:42 | vote | accept | Sam | ||
| Feb 4, 2019 at 19:29 | history | edited | Sam | CC BY-SA 4.0 |
added 32 characters in body
|
| Feb 4, 2019 at 6:00 | history | tweeted | twitter.com/StackStats/status/1092301850823544832 | ||
| Feb 3, 2019 at 17:17 | answer | added | Demetri Pananos | timeline score: 4 | |
| Feb 3, 2019 at 13:51 | history | edited | kjetil b halvorsen♦ |
edited tags
|
|
| Feb 1, 2019 at 0:59 | history | edited | Sam |
edited tags
|
|
| Feb 1, 2019 at 0:35 | comment | added | Sam | @Ingolifs This is really interesting. thanks for it. i will definitely look into it. | |
| Jan 31, 2019 at 23:53 | comment | added | Ingolifs | One good place to start looking may be churn analysis. Churn analysis deals with why people stop using a service, and usage patterns (including login patterns) are a key part of that. Some of the techniques used in such analyses may be of use to you. r-bloggers.com/churn-analysis-part-1-model-selection towardsdatascience.com/… | |
| Jan 31, 2019 at 23:30 | history | edited | Sam | CC BY-SA 4.0 |
added 34 characters in body
|
| Jan 31, 2019 at 23:21 | history | edited | Sam | CC BY-SA 4.0 |
added 157 characters in body
|
| Jan 31, 2019 at 23:12 | comment | added | StatsStudent | If you're privy to network traffic statistics or load, that may be helpful too. | |
| Jan 31, 2019 at 23:12 | comment | added | Sam | Let us continue this discussion in chat. | |
| Jan 31, 2019 at 23:09 | comment | added | StatsStudent | OK. Then it looks like you'll need to use just the limited number of variables that you have. You can try running the algorithms separately for each user or your model might perform better by running the algorithm on the entire dataset and then making individual predictions (for example by treating User as a random effect in you use a regression modelling approach). Regardless it would be imperative to split your data into training and validation datasets, being sure to select the algorithm that leads to the smallest prediction error. | |
| Jan 31, 2019 at 23:06 | comment | added | Sam | I did look into this post which is quite similar to mine but it has more information where as mine doesnot. stats.stackexchange.com/questions/326884/… | |
| Jan 31, 2019 at 23:03 | comment | added | Sam | nope. No IP or even logout time or Amount of time spend etc. nothing. | |
| Jan 31, 2019 at 23:00 | comment | added | StatsStudent | You don't have IP address that can then be used for geolocating users? | |
| Jan 31, 2019 at 22:59 | comment | added | Sam | @StatsStudent No. thats the biggest issue. I only have what is shown in the table above. | |
| Jan 31, 2019 at 22:58 | comment | added | StatsStudent | It seems like you are leaving a lot of data on the table that could be useful to you. I'd imagine you'd have some basic demographics of users that could be powerful predictors. For example, time-zone, age, gender, etc. | |
| Jan 31, 2019 at 22:55 | review | First posts | |||
| Feb 1, 2019 at 0:00 | |||||
| Jan 31, 2019 at 22:54 | history | asked | Sam | CC BY-SA 4.0 |