I have created a dataframe that contains a year's worth of data for package downloads in my country from CRAN and I now want to go about building a network of usage (I am not sure if network analysis is the correct term here).
There is a daily unique id assigned to each IP address (ip_id) so this combined with the day creates an overall unique ID for the set. I want to take each one of these instances and add it to a network so you will be able to see the cross usage between packages downloaded in these instances.
Each row in my set contains a value in ip_id and one package name (package). So if you downloaded 5 packages on this day - you will be 5 rows in the data.
I am assuming I will have to create some matrix that has a count of values that if car and crayon are selected they both get a one in their own match points on the diagonal and then also a 1 in the point where they correspond.
And then only after having created this I would be able to visualize it?
Added: 1 See example of the data. So sourcetools, shiny, Cairo, htmlwidgets and DT would be one set for ip_id 2904 which means shiny x shiny should get 1 shiny x Cairo should get 1 and so and so forth.
If I can get this grid sorted out I should be able to visualize it.