Summer 2014 Week One Tasks

[aflaxman]

During june 16-22 week:

Hard work:

• [x] Extract json data into usable format
• [x] Classify users and tweets
• [x] See if sentiment persists over time at the city level (and individual level if possible)

Easy work:

• [x] Add ADF to github repo
• [ ] Find iSchool text analysis platform

Medium work:

• [ ] Computer methods writeup
• [ ] Reproducibility project experience writeup

[aflaxman]

For second week:

• [ ] finish refactoring for classification and then check persistence
• [x] move writeups to repo (in markdown)
• [x] add iSchool links to this discussion
• [x] check off checkboxes.

Also to do:

• [ ] see how classification performs on 2014 data

[bpcb]

iSchool -- Cecelia Aragon https://faculty.washington.edu/aragon/

Relevant publication: Automating Large-Scale Annotation for Analysis of Social Media Content http://research.microsoft.com/en-us/um/people/shliu/tasmc/paper/Automating.pdf

[aflaxman]

• [ ] lit review
• [ ] draft results section

[bpcb]

Not promising results...here are two figures comparing data from 2009 and 2014 by city and by county:

[aflaxman]

Try one where each dot is an individual.