3
$\begingroup$

I am currently trying to find clusters in a data set that looks like this:

         Dienstag 19 Mittwoch 20 Donnerstag 21 Freitag 22 Montag 25 Dienstag 26 Donnerstag 28
 [1,]           0           0             0          0         0           0            NA
 [2,]           0           0             0          0         0           0            NA
 [3,]           0           0             0          0         0           0            NA
 [4,]           0           0             0          0         1           0            NA
 [5,]           1           0             1          1         1           1            NA
 [6,]           0           0             0          0         0           0            NA
 [7,]           4           0             1          0         2           1            NA
 [8,]           0           1             2          1         0           2            NA
 [9,]           0           0             1          0         0           0            NA
[10,]           1           0             0          0         0           1             0
[11,]           2           0             1          0         0           5             0
[12,]           1           0             0          0         0           1             1
[13,]           0           1             0          0         0           0             0
[14,]           0           0             1          0         4           1             0

It corresponds at the counting of times a user used a service given the day and the hour.

I want to find pattern/clusters that relate the usage of a weekday with the hour, but I dont know how to manage it. Lets say for example that from monday to friday the usage is mostly concentrated between 5 and 16 hour, or that there are two blocks of mayor usage along the day.

The final objective of this is to predict the hours where the usage will be concentrated in the next days, for applying a control algorithm.

It would really be helpful if you could give me some suggestions about methods or procedures to do that.

Thanks

$\endgroup$
  • 7
    $\begingroup$ A very underrated method: plot that stuff. More specifically, plot the average counts over all weekdays for each hour between 7 and 20 - maybe you see spikes during lunchbreak or something similar. If you do, confirm this is a phenomenon present for all weekdays, not just one or two. And then you have your pattern. $\endgroup$ – Nameless Jul 3 '13 at 9:44
1
$\begingroup$

To add on to Nameless's response, you could try to use this calendar heat map to plot your raw counts in R. http://www.inside-r.org/packages/cran/makeR/docs/calendarHeat

It might make it easier to see weekend/weekday patterns, etc.

$\endgroup$
  • $\begingroup$ I already have a graphical representation of the data and can see that the usage is concentrated in between 8 and 17 for one case but I have many other cases, I am trying to find if there is any method to accomplish this without me having to manually idenditfy the pattern for every case. $\endgroup$ – Tarigarma Jul 3 '13 at 14:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.