//EmoTweet - A real-time emotional analysis of tweets
Six images as a response to six basic emotions – happiness, anger, fear, surprise, sadness, and disgust (left to right, top to bottom)
What?
EmoTweet based on the Synesketch API, sources live tweets of users, senses emotions and creatively visualizes them. The six basic emotions - happiness, anger, fear, surprise, sadness, and disgust are depicted in the above picture, a combination of these six basic emotions at various scales make up a final visualization of the tweet. Currently EmoTweet is a desktop Java based software and uses the Twitter API to source user tweets.
Why?
Twitter is used by people to connect with friends, relatives and co-workers. As such twitter has a huge information flow - and most of it is day to day activities and their associated experiences. I think it's an interesting challenge to analyze the huge database of personal expressions in terms of emotional visualizations. Apart from being an immense asset in studying the daily patterns, it helps us in understanding human psychology in the context of tweets
How?
The Synesketch API forms the core of EmoTweet, the code serves as a medium between words, emotions, and images. A result of a research that spreads out through several diverse fields – from natural language processing techniques based on WordNet, across Ekman's research of emotions, to color psychology, visual design, data visualizations, and affective computing. Graphics were done by Processing, a great tool for programming visual art. Twitter API was used to source real time tweets.
Textual emotion recognition is based on the lexicon of emotion-related words, where each word is associated with a vector of weights for each emotion type (happiness, sadness, fear, anger, disgust, surprise). The emotion sensing algorithm itself uses the Lexicon, and by several simple heuristics (usage of upper cases, exclamation mark, emoticons etc.) estimates the emotional content of the sentence.
And..
Currently I am working on porting EmoTweet online thus ensuring an open access to the entire community. Also the approach that the Synesketch API takes is that of words and not that of context - I am trying to make affective computing, a context sensitive viable reality for EmoTweet!
//EmoTweet - A real-time emotional analysis of tweets
Six images as a response to six basic emotions – happiness, anger, fear, surprise, sadness, and disgust (left to right, top to bottom)
What?
EmoTweet based on the Synesketch API, sources live tweets of users, senses emotions and creatively visualizes them. The six basic emotions - happiness, anger, fear, surprise, sadness, and disgust are depicted in the above picture, a combination of these six basic emotions at various scales make up a final visualization of the tweet. Currently EmoTweet is a desktop Java based software and uses the Twitter API to source user tweets.
Why?
Twitter is used by people to connect with friends, relatives and co-workers. As such twitter has a huge information flow - and most of it is day to day activities and their associated experiences. I think it's an interesting challenge to analyze the huge database of personal expressions in terms of emotional visualizations. Apart from being an immense asset in studying the daily patterns, it helps us in understanding human psychology in the context of tweets
How?
The Synesketch API forms the core of EmoTweet, the code serves as a medium between words, emotions, and images. A result of a research that spreads out through several diverse fields – from natural language processing techniques based on WordNet, across Ekman's research of emotions, to color psychology, visual design, data visualizations, and affective computing. Graphics were done by Processing, a great tool for programming visual art. Twitter API was used to source real time tweets.
Textual emotion recognition is based on the lexicon of emotion-related words, where each word is associated with a vector of weights for each emotion type (happiness, sadness, fear, anger, disgust, surprise). The emotion sensing algorithm itself uses the Lexicon, and by several simple heuristics (usage of upper cases, exclamation mark, emoticons etc.) estimates the emotional content of the sentence.
And..
Currently I am working on porting EmoTweet online thus ensuring an open access to the entire community. Also the approach that the Synesketch API takes is that of words and not that of context - I am trying to make affective computing, a context sensitive viable reality for EmoTweet!
