Signal Spotlight provides a real-time overview of the emotional state of Signal search results. Using Signal Spotlight, Signal users can better understand the prevalence and drivers of emotions and what is happening in real-time.
Spotlight taps into the results from Signal search criteria across many data sources to better understand the prevalence and drivers of emotions. For example, during an incident or event, an important attribute is how people are feeling about what has happened and how the emotional state may be changing real-time as that incident/event unfolds.
The Spotlight underlying technology uses a large vocabulary of emotion terms that were compiled from multiple sources, including the ANEW and LIWC corpora, and a list of moods from LiveJournal. In addition, a crowdsourcing task was run to organise these terms against Parrott's hierarchy of emotions. The emotions are colour-coded using a dataset of affective norms provided by the Center for Reading Research at Ghent University.
Spotlight leverages technology and research undertaken by the Language and Social Computing team in the Digital Economy Programme of CSIRO's Digital Productivity Flagship and originally developed as a joint project between computer scientists at CSIRO and mental health researchers at The Black Dog Institute.
Reference
Milne, D., Paris, C., Christensen, H., Batterham, P. and O'Dea, B. (2015) We Feel: Taking the emotional pulse of the world. In the Proceedings of the 19th Triennial Congress of the International Ergonomics Association (IEA 2015), Melbourne, Victoria, Australia, August 2015.