Extracting relevant knowledge for the detection of sarcasm and nastiness in the social web | Publicación