The popularity of online social platforms has also determined the emergence of violent and abusive behaviors reflecting real life issues into the digital arena. Cyberbullying, Internet banging, pedopornography, sexting are examples of these behaviors, as witnessed in the social media environments. Several studies have shown how to approximately detect those behaviors by analyzing the social interactions and in particular the content of the exchanged messages. The features considered in the models basically include detection of o ensive language through NLP techniques and vocabularies, social network structural measures and, if available, user context information. Our goal is to investigate those users who adopt offensive language and hate speech in Twitter by analyzing their profile pictures. Results show that violent people smile less and they are dominating by anger, fear and sadness.
Do violent people smile? Social Media analysis of their profile pictures
C. Lucchese;S. Orlando
2018-01-01
Abstract
The popularity of online social platforms has also determined the emergence of violent and abusive behaviors reflecting real life issues into the digital arena. Cyberbullying, Internet banging, pedopornography, sexting are examples of these behaviors, as witnessed in the social media environments. Several studies have shown how to approximately detect those behaviors by analyzing the social interactions and in particular the content of the exchanged messages. The features considered in the models basically include detection of o ensive language through NLP techniques and vocabularies, social network structural measures and, if available, user context information. Our goal is to investigate those users who adopt offensive language and hate speech in Twitter by analyzing their profile pictures. Results show that violent people smile less and they are dominating by anger, fear and sadness.File | Dimensione | Formato | |
---|---|---|---|
mainWWW.pdf
non disponibili
Tipologia:
Documento in Pre-print
Licenza:
Accesso chiuso-personale
Dimensione
433 kB
Formato
Adobe PDF
|
433 kB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.