Goals

• Learn about using word lists to determine sentiment of written content.
• Get an idea about spreading in networks.

Reading

• Dodds PS, Harris KD, Kloumann IM, Bliss CA, Danforth CM (2011) Temporal Patterns of Happiness and Information in a Global Social Network: Hedonometrics and Twitter. PLoS ONE 6(12): e26752. doi:10.1371/journal.pone.0026752 [TPHIGSN]
• Catherine A. Bliss, Isabel M. Kloumann, Kameron Decker Harris, Christopher M. Danforth, Peter Sheridan Dodds (2012). Twitter reciprocal reply networks exhibit assortativity with respect to happiness. Journal of Computational Science 3 p388–397. [TRRNEAWRTH]
• Johan Bollen, Bruno Gonçalves, Guangchen Ruan, & Huina Mao (2011) Happiness is assortative in online social networks. http://arxiv.org/pdf/1103.0784v1.pdf [HIAIOSN]

Exercises

1) Twitter network assortativity.

• Read RRNEAWRTH. Explain the main findings in your own words. Use about 1000 characters.
• Read HIAIOSN. Explain the main findings in your own words. Use about 1000 characters.
• Compare and contrast the two papers. What are the main shared findings? Are there any differences.
• Measure assortativity on your own.

• Start from Twitter user Brian Keegan (@bkeegan). Calculate average sentiment for his most recent tweets, using the word-list from TPHIGSN.
• Calculate sentiment for 50 of his reciprocal followers.
• Calculate sentiment for 50 of his non-reciprocal followers.
• Is the sentiment of the reciprocal followers closer to Brian’s sentiment than for the non-reciprocal followers?
• How do your results line up with the findings you read about earlier?

 2) Take the last steps to get the bot ready for the final project

• Start following the accounts from San Francisco that follow other people in the class. The list can be found here. Make sure you download the list often, as it will keep growing.
• For every human follower you gain, use the network and follow all of their reciprocal followers (people they follow and who follows them back) Note: Do not include the bots in the class in the reciprocal following step!  You can see a list of the bots in the class here.
• Generate one original tweet per day. You may choose your own method, e.g. grabbing content from popular web-pages. Or follow some of the strategies that Claire outlined during the lecture today.
• For every human you attempt to follow, store the entire user profie for later analysis (see Lecture 8). Keep track of who follows you back and who doesn't. IMPORTANT (this one is necessary for the final project).