Nowadays, big volumes of User-Generated Content (UGC) spread across various kinds of social media. In microblogging, UCG can be generated in the form of ‘newsworthy’ posts, i.e., related to information that has a public utility for the people. In this context, being the UGC diffused without almost any traditional form of trusted external control, the possibility of incurring in possible fake news is far from remote. For this reason, several approaches for fake news detection in microblogging have been proposed up to now, mostly based on machine learning techniques. In this paper, an ongoing work based on the use of the Multi-Criteria Decision Making (MCDM) paradigm to detect fake news is proposed. The aim is to reduce data dependency in building the model, and to have flexible control over the choices behind the fake news detection process.
De Grandis, M., Pasi, G., Viviani, M. (2019). Fake News Detection in Microblogging Through Quantifier-Guided Aggregation. In MDAI 2019: The 16th International Conference on Modeling Decisions for Artificial Intelligence - Milan, Italy, September 4 - 6, 2019 (pp.64-76). Springer Verlag [10.1007/978-3-030-26773-5_6].
Fake News Detection in Microblogging Through Quantifier-Guided Aggregation
Pasi, Gabriella;Viviani, Marco
2019
Abstract
Nowadays, big volumes of User-Generated Content (UGC) spread across various kinds of social media. In microblogging, UCG can be generated in the form of ‘newsworthy’ posts, i.e., related to information that has a public utility for the people. In this context, being the UGC diffused without almost any traditional form of trusted external control, the possibility of incurring in possible fake news is far from remote. For this reason, several approaches for fake news detection in microblogging have been proposed up to now, mostly based on machine learning techniques. In this paper, an ongoing work based on the use of the Multi-Criteria Decision Making (MCDM) paradigm to detect fake news is proposed. The aim is to reduce data dependency in building the model, and to have flexible control over the choices behind the fake news detection process.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
