A Topic-Agnostic Approach for Identifying Fake News Pages

Abstract

Fake news and misinformation have been increasingly used to manipulate popular opinion and influence political processes. To better understand fake news, how they are propagated, and how to counter their effect, it is necessary to first identify them. Recently, approaches have been proposed to automatically classify articles as fake based on their content. An important challenge for these approaches comes from the dynamic nature of news: as new political events are covered, topics and discourse constantly change and thus, a classifier trained using content from articles published at a given time is likely to become ineffective in the future. To address this challenge, we propose a topic-agnostic (TAG) classification strategy that uses linguistic and web-markup features to identify fake news pages. We report experimental results using multiple data sets which show that our approach attains high accuracy in the identification of fake news, even as topics evolve over time.

Fake News Early Detection

“…Style is a set of self-defined [non-latent] machine learning features that can represent fake news and differentiate it from the truth [65]. For example, such style features can be word-level statistics based on TF-IDF, n-grams and/or LIWC features [6,40,41], and rewrite-rule statistics based on TF-IDF [40]. Though these style features can be comprehensive in detecting fake news, their selection or extraction is driven by experience that is rarely supported by fundamental theories across disciplines.…”

Uma Introdução ao Combate Automático às Fake News em Redes Sociais Virtuais

“…O segundo grupo, entretanto, tem uma definição mais genérica. Para este segmento, as Fake News são todas as notícias falsas, independente da sua natureza intencional [Sharma et al 2019] [Castelo et al 2019] [Ajao et al 2019]. Inclusive, consideramse como Fake News outros tipos de notícia, como, por exemplo, Rumor.…”

“…Além disso, os referidos trabalhos são brevemente descritos, podendo seus detalhes serem consultados através das respectivas referências: T1) A Topic-Agnostic Approach for Identifying Fake News Pages [Castelo et al 2019]: O trabalho propõe um topic-agnostic (TAG) classificador que usa dados linquísticos e Web-Markup (padrões de layout das páginas) para detectar Fake News. Assim, ao invés de usar o bag of words, o trabalho explora as topic-agnostic, incluindo características morfológicas, psicológicas e de legibilidade que são comuns em Fake News.…”

On the Detection of False Information: From Rumors to Fake News

“…It is worthy to mention at this point that a lot of work in the literature did not focus on the fine-grained types of false information, and they used the terms fake or unreliable instead (e.g., fake vs real, as a binary classification task). The work in this line varies in terms of the used models, from simple handcrafted based systems to models based on deep learning [165,186,34,201].…”

“…This is usually done together with adding misleading terms that can have a negative or positive impact on the readers' emotions. Previous work [186,34,201] have discarded the sequential order of events in fake news articles. In this section we propose a model that takes into account the affective changes in texts to detect fake news.…”