—-BAKER—-
Please add the following to the batter:
Q Data Set (.csv and .xls): https:// anonfile.com/mei9s5d9ba/q_data_set_1.5.zip
There is a lot you can do with analytics and visualization if you have a workable data set, where everything is broken out into a nice database.
Lot's of versions of the data are floating around for people to use, but I haven't come across something like this so I made it.
I also took the time to tag each line with associated signatures from that post, which can be useful for mapping out relationships (think chord charts, network maps, venn diagrams).
If I have the time and figure out an efficient way, I'll continue updating and tagging lines with relevant data points (people, places).
The version already there is a bit outdated, this is an updated file with new info on it.
Previous link was wrong, here is the correct link:
https:// anonfile.com/meI9s5d9ba/Q_Data_Set_1.5.zip
Unfortunately I don't do much data analytics, so although I'm aware of the potential, I'm pretty slow and clumsy with it. I'm going to work on a Venn diagram showing all the posts that fall under the various signatures, I think that could be useful.
Ultimately what would really unlock this is if I could accomplish the next step, adding attributes to each row such as:
>Person
>Place
>Organization
>If the post contains [] markers
>etc.
Then, you could view the relationships between these attributes, based on them being tied together by other attributes, such as the same post, or the same signature, etc.
I'm not sure how much more I'll reasonably be able to do alone to improve this, but I think it's important we have something like this available for any anon who has analytics skills.
>And as far as your document, can you just add separate columns for each row to add those attributes?
Yes, and then fill in that cell for any row that has a relevant data point. Problem is you need a way to normalize both "What's north of SK?" and "What does NK stand for?" into "North Korea." A certain amount of that could maybe be automated for big keywords, but much of it would have to be manual, I think.