Sentiment Analysis - #WinterIsHere
My MSc dissertation was a team project exploring sentiment analysis - using Twitter’s API to source our data, we applied our method to the content of users’ tweets in order to assess the global opinion on a given topic.
As proof-of-concept we carried out case studies as examples of different applications of the tool. One such study was the viewer reaction to character deaths in the TV series Game of Thrones.
Character deaths typically spawn a strong online reaction - negative for a fan-favourite and positive for a dislikeable character - which meant a show infamous for killing off characters unexpectedly was an excellent source of data.
The chart displays characters organised in order of death; sized and color-coded according to the scale of the online reaction. It also served as a method of quickly navigating to a character's place on the timeline.
All 200+ dead characters are included with their allegiance and cause of death, but only characters that sparked a lot of discussion received in-depth analysis.
Allegiances were represented by icons inspired by sigils from the show and books.