Profiling Bots and Fake News Spreaders at PAN'19 and PAN'20 : Bots and Gender Profiling 2019, Profiling Fake News Spreaders on Twitter 2020

• Published in: 2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA) pp. 626 - 630
• Main Author: Pizarro, Juan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2020
• Subjects: Author Profiling; Birds; Blogs; Bot identification; Fake News; Gender Identification; Social networking (online); Support vector machines; Task analysis; Training; Tuning; Twitter
• DOI: 10.1109/DSAA49011.2020.00088

This paper synthesizes our participation in the CLEF conference 2020 regarding the Profiling Fake News Spreaders on Twitter task, organized at the PAN lab on digital text forensics and stylometry. Our team obtained one of the two best performing results with an average accuracy of 0.7775 -0.7350 for English and 0.8200 for Spanish. In summary, we submit a Support Vector Machine (SVM) classifier trained with character and word n-gram to determine whether the author of a tweet feed is keen to be a spreader of fake news. Besides, we present our results as the best performing team in Bots and Gender Profiling 2019 with an average accuracy of 0.8805. Its goal was to determine if the author of a tweet feed was a bot or human, followed by identifying her/his gender in the case of humans.