Shiny based tool for discrete temporal events visualization

131 visningar
Hoppa till det första olästa meddelandet

Omri Mendels

oläst,
2 feb. 2018 13:49:302018-02-02
till Israel R User Group
Hi,

I recently finished working on a tool that might be relevant to you if you're working with temporal discrete events (or intervals).
It fills a need that most time series packages don't which is dealing with discrete time series rather than continuous.

For example, you could see patterns of co-occurring events, consecutive events, differences in distributions between sessions in your data and more.

Could be useful if you're into analyzing/modeling:
- Human behavior
- Cyber security
- IoT
- Any other discrete events data that has a value, start and end times attributes


An example (based on the ADL dataset in UCI repo):
Patterns of consecutive events (Markov chain like):
Consecutive

Patterns of co-occurring events:
Consecutive

Events timeline:
Timeline


Any feedback is mostly welcome!
Omri


Jonathan Rosenblatt

oläst,
4 feb. 2018 03:59:212018-02-04
till israel-r-user-group
Very cool!


--
You received this message because you are subscribed to the Google Groups "Israel R User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to israel-r-user-group+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--
--
Jonathan Rosenblatt
Dept. of Industrial Engineering and Management
Ben Gurion University of the Negev

Svara alla
Svara författaren
Vidarebefordra
0 nya meddelanden