I am analyzing an aws log file with http request logs, and I need to predict the expected load (number of requests) in the next minute. However, I see there are spans of times that doesn't have any logs. In this kind of case, do I assume that loads during those times were just 0, or do I need to do some sort of interpolation?
time load
-----------------------------------
2018-11-07 09:45:00 40
2018-11-07 09:46:00 45
2018-11-07 09:47:00 34
2018-11-07 09:48:00 56
and then no logs for the next 2 hours and then again:
time load
-----------------------------------
2018-11-07 11:50:00 54
2018-11-07 11:51:00 34
2018-11-07 11:52:00 23
2018-11-07 11:53:00 21
Let's say when I read this file to a pandas dataframe for my prediction model, do I fill in all the minutes for those 2 hours with 0? Or are there better intelligent ways of dealing with this sort of situations?