amp5 / emoji_sa Goto Github PK

• define all variables and determine what vars you need in analysis

• save only needed vars into data frame for further analysis

deadline for 2/8/17
decide whether I need to de-identify twitter data

Decided to use issues instead. Figure out how to delete.

• narrow in on observations of interest (US only)
• create new vars that are functions of existing vars

and then add that into github
deadline - 2/8/17

consider putting them in one file?

• save only id and username into a new df with '_' as break. ie. 89137491_username123
• concatenate id and username into one col
• get unique values of new one col
• separate col based on '_'
• store this file in .gitignore so this potentially identifying info is not shared on github but I still have access to it
• then remove username from raw data and upload raw data to github
• begin