visualizing the crisis

VISUALIZING THE CRISIS  is a research project that explores the potential of text mining methodsapplied to geo-localized user generated content in order to extract and analyze the emotionalreaction of Italian people to the current political crisis.The project focused on building a platform that harvests and integrates real-time datastreams coming from geo-localized user generated content and then applies text miningprocesses in order to extract users’ emotional reactions and plot them on a map.The project started analyzing users’ emotional temperature in the two major Italian cities(Rome and Milan) in November 2011, when Silvio Berlusconi lost his parliamentary majorityand Mario Monti was nominated Prime Minister.The idea behind the project was to create a framework to continuously listen and monitorurban actors’ emotional reactions and enhance their sense of agency and their responsivenessduring the current period of political and economical transformations in Italy.The paper introduces the theoretical background and the technological characteristics of theplatform and presents a short selection of interesting patterns and urban stories we cameacross during the early phase of the study. Some relevant images are presented and discussedas well.

- – -
We’re presenting a full paper at the Cumulus conference (http://cumulushelsinki2012.org/).

VISUALIZING THE CRISIS is developed with Luca Simeone, Paolo Patelli  and Salvatore Iaconesi.
The paper preview can be seen here.

The video log of the extractions and sentiment analysis we’re using for the city of Milan can be found here.
The video represents only a plotting of the different geo-localized UGC contribution with the relative emotion.
We’re currently experimenting on visualization, in order to provide a meaningful and intuitive visual feedback that could act as a tool that enlightens spatial clusters and temporal relationships to enable proper interpretations of the phenomena.

(Figure 1)
Visualizing the crises. Geo-located UGC and related emotions shared within the city of Rome on January 18th, 2012 – visual experiment 1 (single contributions plotting)

- – -

(Figure 2)
Visualizing the crises. Geo-located UGC and related emotions shared within the city of Rome on January 18th, 2012 – visual experiment 2 (interpolation of average sentiment heatmap)

- – -
interpretations and methodology
Below there are two specific examples of queries performed during the research projectthrough the Visualizing the crisis platform.

January 18, 2012
- Event: Mario Monti visits David Cameron in London and discusses his goals: labour market reform,flexsecurity and the Danish model along with the fight against tax evasion.
- Sociopolitical keywords related to the event: Monti, Cameron, Lavoro, Occupazione, Licenziamenti, Legge, Evasione, Danimarca,Flexsecurity, Articolo 18, Riforme.
- Geographical outcomes emerging from the platform Visualizing the crisis:
Milan: We noticed a high concentration of positive feelings around the very center of the cityof Milan (where banks and financial companies are located) along with other more sporadicspikes around other business and media hubs (Porta Garibaldi, Cologno Monzese, CorsoSempione/Certosa, Rogoredo…). We could then notice a progressive shifting towards morenegative feelings in the industrial clusters around the city (Sesto San Giovanni, Rho/Pero,San Donato…). Even more interestingly though, we could identify an averagely positivereaction around lower and middle class residential neighborhoods probably suggestingevidence of the unexpected hope the government change brought even to the groups ofpeople more affected by the hardness of the economical measures.

January 23, 2012
- Event: Nation wide strike of taxi drivers causes many traffic problems in major Italian cities. Thedrivers protest against a liberalization of the market proposed by the government as one ofthe key reforms in order to contain the effects of the economical crisis.
- Sociopolitical keywords related to the event: Monti, Governo, Riforme, Liberalizzazioni, Decreto, Categorie, Licenze, Tassisti, Taxi,Sciopero, Traffico, Servizio Pubblico.
- Geographical outcomes emerging from the platform Visualizing the crisis:
Rome: We noticed an average negative reaction, evenly spread across the city, slowly fadingto more neutral feelings from the center towards the outskirts.Higher concentrations along tourist routes and city landmarks suggest disappointmentespecially among visitors and taxi users, as if the negative feelings were directed towards thestrike itself other than the government and the reforms. Furthermore, the researchhighlighted the usage of specific ironic hashtags (e.g. #menotaxipertutti) related to negativeUGCs and addressing the strike as a source of disaffection.In these two examples, Visualizing the crisis helped us to understand how cities react topolitical and economic events, relating citizens’ emotions to specific sociopoliticalcircumstances and to specific spatial and temporal coordinates.An important component of the platform is the visualization engine. When the appropriatedata are presented together in the right context, they can help the reader to understand thesituation they are depicting or even discover new relationships that have been previouslyhidden in the data.Two early visual representations (based on the city of Rome) are enclosed below.

About giorgia lupi

1981 I am an architect that never built any house (luckily). I work with information, designing, researching and drawing a lot. I am co-founder ad design director at Accurat (www.accurat.it) I am a PhD student at Milan Politecnico (www.densitydesign.org)

2 comments

  1. What a data of un-ambiguity and preserveness
    of valuable familiarity about unexpected feelings.

  2. Great information. Lucky me I came across your blog by accident (stumbleupon).
    I’ve bookmarked it for later!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 3,392 other followers

%d bloggers like this: