This is the logic used to produce the JSON files
user@user:~/facebook$ STARTDAY="2017-12-07" ENDDAY="2017-12-20" DEBUG=* mongo-scripts/silver.js
silver [+] All the impressions are 12670 +10ms
silver Opening and eventually overwritting 'impressions - 13 days.json' +903ms
silver [+] Unique posts are 2348 +3s
silver Opening and eventually overwritting 'posts - 13 days.json' +294ms
silver Done! files saved +114ms
This file import the product of Facebook API to get insight on the actual amount of posts publiched by the sources, the import script is apiimport.js
The file gauss.json analyze the visualizationDiff value, rounding the value in hours. It could help to study the competitiion among the same post, as the log show:
user@user:~/experiments-data/wto$ DEBUG=* ../competition.js
competition Opening file impressions - 13 days.json +0ms
competition Writings gauss.json, 12670 visualization object: done! +825ms
it is based on the impressions.
This compare the API collected post vs the timeline observed posts