In this paper, we summarize our participation in the task of “Profiling Hate Speech Spreaders on Twitter” at the PAN@CLEF Conference 2021. Our models obtained an average accuracy of 76% (79% for Spanish and 73% for English). For English, we used a Linear Support Vector Machine with tf-idf features on noun chunk level, while for Spanish we used a Ridge Classifier with simple counts on noun chunk level. Both classifiers were fed with additional features obtained from a Convolutional Neural Network.    «

In this paper, we summarize our participation in the task of “Profiling Hate Speech Spreaders on Twitter” at the PAN@CLEF Conference 2021. Our models obtained an average accuracy of 76% (79% for Spanish and 73% for English). For English, we used a Linear Support Vector Machine with tf-idf features on noun chunk level, while for Spanish we used a Ridge Classifier with simple counts on noun chunk level. Both classifiers were fed with additional features obtained from a Convolutional Neural Network...    »