Browsing by Author "Meiselwitz, G"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    An Empirical Analysis of Rumor Detection on Microblogs with Recurrent Neural Networks
    (SPRINGER INTERNATIONAL PUBLISHING AG, 2019) Bugueno, Margarita; Sepulveda Villalobos Gabriel Andres; Mendoza Rocha Marcelo Gabriel; Meiselwitz, G
    The popularity of microblogging websites makes them important for information dissemination. The diffusion of large volumes of fake or unverified information could emerge and spread producing damage. Due to the ever-increasing volume of data and the nature of complex diffusion, automatic rumor detection is a very challenging task. Supervised classification and other approaches have been widely used to identify rumors in social media posts. However, despite achieving competitive results, only a few studies have delved into the nature of the problem itself in order to identify key empirical factors that allow defining both the baseline models and their performance. In this work, we learn discriminative features from tweets content and propagation trees by following their sequential propagation structure. To do this we study the performance of a number of architectures based on recursive neural networks conditioning for rumor detection. In addition, to ingest tweets into each network, we study the effect of two different word embeddings schemes: Glove and Google news skip-grams. Results on the Twitter16 dataset show that model performance depends on many empirical factors and that some specific experimental configurations consistently drive to better results.
  • Thumbnail Image
    Item
    Claim Behavior over Time in Twitter
    (SPRINGER INTERNATIONAL PUBLISHING AG, 2019) Weiss, Fernanda; Espinoza, Ignacio; Hurtado Gonzalez Julio Andres; Mendoza Rocha Marcelo Gabriel; Meiselwitz, G
    Social media is the primary source of information for many people around the world, not only to know about their families and friends but also to read about news and trends in different areas of interest. Fake News or rumors can generate big problems of misinformation, being able to change the mindset of a large group of people concerning a specific topic. Many companies and researchers have put their efforts into detecting these rumors with machine learning algorithms creating reports of the influence of these "news" in social media (https://www.knightfoundation.org/reports/disinformation-fake-news-and-influence-campaigns-on-twitter). Only a few studies have been made in detecting rumors in real-time, considering the first hours of propagation. In this work, we study the spread of a claim, analyzing different characteristics and how propagation patterns behave in time. Experiments show that rumors have different behaviours that can be used to classify them within the first hours of propagation.
  • Thumbnail Image
    Item
    From Belief in Conspiracy Theories to Trust in Others: Which Factors Influence Exposure, Believing and Sharing Fake News
    (SPRINGER INTERNATIONAL PUBLISHING AG, 2019) Halpern, Daniel; Valenzuela, Sebastián; Katz, James; Miranda, Juan Pablo; Meiselwitz, G
    Drawing on social-psychological and political research, we offer a theoretical model that explains how people become exposed to fake news, come to believe in them and then share them with their contacts. Using two waves of a nationally representative sample of Chileans with internet access, we pinpoint the relevant causal factors. Analysis of the panel data indicate that three groups of variables largely explain these phenomena: (1) Personal and psychological factors such as belief in conspiracy theories, trust in others, education and gender; (2) Frequency and specific uses of social media; and (3) Political views and online activism. Importantly, personal and political-psychological factors are more relevant in explaining this behavior than specific uses of social media.