Persuasive Natural Language Processing: from Web advertising to political speeches and virality in social networks
ILC-CNR - Aula Seminari IBF SG 5
While there has been a vast amount of research and applicative works on understanding the semantics of a text, little has been done - so far - on understanding how such texts can be used to modify human behavior.
Trivially, reactions to textual content are connected to language use. While such reactions can vary across domains, audiences and scenarios, some underlying phenomena can be modeled in a formal way and analyzed through Natural Language Processing tools.
Using various examples (from Web ads to political speeches and contents posted on social networks), I will address the topic of language persuasiveness, i.e. how language can be automatically analyzed (and possibly modified) to bring about various audience reactions.