Felicitt` a: Mussa, C. Bosco, V. Patti Visualizing and Estimating - - PowerPoint PPT Presentation

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V. Mussa,
• C. Bosco, V. Patti and G. Ruffo

Dipartimento di Informatica, Universit` a di Torino

December 3, 2013

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Outline

Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Analyzing and visualizing data from social media

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Social media users generated contents represent the human behavior in everyday life, but ... how to analyze data from social media? how to connect quantitative research with theories that would qualitatively explain the observed and modeled phenomena? how to visualize data from social media? statistical tools, machine learning techniques, large-scale network analysis and natural language modeling are hard to be applied by many skilled sociologists and communication scientists that do not want to deal directly with raw data or abstract models.

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Felicitt` a

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

http://www.felicitta.net

An online platform for estimating happiness in Italian cities which provides visualization techniques to interactively explore the result of sentiment analysis performed over geotagged Tweets

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Felicitt` a is a fully implemented visualization system to estimate the level of happiness in a given geographical area based on geotagged Tweets, overlaying different analysis engines and visualization techniques a particular instantiation of the system, where a sentiment analysis engine for detecting Italian Tweets’ sentiment polarity has been developed and employed in order to estimate happiness in Italian cities and regions

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Architecture

The Felicitt` a framework includes: geotagged information retrieval from social media APIs data enrichment with socio demographic information from statistics and other sources (e.g. ISTAT, Italian National Institute of Statistics) analysis of data producing an estimation of the general sentiment at different geographical and temporal scales visualization of analysis results

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

The Felicitt` a framework

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Sentiment analysis

The analysis engine performs three main steps: pre-parsing: Tweets collection and cleaning by deletion or substitution of emoticons, links, mentions

• f other users and redundant punctuation

parsing: Tweets morpho–syntactic analysis and lemmatization (by Freeling) analysis: match of content words with MultiWordNet and WordNet–Affect entries, aggregation of single words polarities, aggregation according to geolocation of Tweets polarities

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Visualization

Visualization techniques are adopted to support researchers and practitioners to explore the data about happiness in Italian cities. Three main views are implemented for describe happiness in regions, cities and top ten happiest places. An interactive interface where users can request for giving more contextual and quantitative details on maps, plots, tag clouds and other charts.

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Happiness in the cities

Felicitt` a shows in the view Citt` a for each day: a map of Italy a round marker for each town, which assumes different colors and sizes according to affective status and amount of posted messages an ordered score of the Italian towns details about the evaluation expressed by the system and Tweets posted there to test the reliability of the evaluation, for each city selected by the user

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

The view Citt` a

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Happiness in the regions

Felicitt` a shows in the view Regions for each day: a map of Italy where the regions are colored according to their affective status a score of the regions from the happiest to the less happy

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

The view Regions

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Top ten happiest places

Felicitt` a shows in the view Top ten for each day: a map showing all the Tweets positioned within the area where they have been posted; by moving on the map, the user can zoom in to find the exact position

• f Tweets on the map and also to read them

each Tweet is represented by a marker colored according to its detected affective polarity

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

The view Top ten

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Diachronic variation of happiness

To observe how the level of happiness varies in time, Felicitt` a offers a quarterly view:

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Diachronic variation of happiness

... and a daily based view

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Clouds for happiness

To observe the relations between happiness and words, Felicitt` a can display tag-clouds, a visual representation useful for quickly perceiving the most prominent terms involved in analyzed Tweets. Tag-cloud of the happiest august day in Turin

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Conclusion and future work

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Our first focus was on the sentiment visualization and summarization, but in order to improve Felicitt` a we are working on other issues, in particular: to develop a sentiment analysis approach which goes beyond the simple lexicon-based one. For this purpose, we are developing a gold standard corpus

• f manually annotated Tweets to be used as a

testbed for evaluation and comparison with other systems. to apply finer grained emotion detection techniques in order to classify Tweets according to different emotions (e.g. Ekman’s basic emotions) or the emotional categories from the Plutchik’s model to provide a sort of geography of emotions.

Felicitt` a: Visualizing and Estimating Happiness in Italian Cities from Geotagged Tweets

• L. Allisio, V.

Mussa,

• C. Bosco, V. Patti

and G. Ruffo Analyzing and visualizing data from social media Felicitt` a

Architecture Sentiment analysis Visualization

Conclusion and future work

Thank you and ...