Online Social Media (OSM) have been substantially transforming the process of spreading news, improving its speed, and reducing barriers toward reaching out to a broad audience. However, OSM are very limited in providing mechanisms to check the credibility of news propagated through their structure. The majority of studies on automatic fake news detection are restricted to English documents, with few works evaluating other languages, and none comparing language-independent characteristics. Moreover, the spreading of deceptive news tends to be a worldwide problem; therefore, this work evaluates textual features that are not tied to a specific language when describing textual data for detecting news. Corpora of news written in American English, Brazilian Portuguese, and Spanish were explored to study complexity, stylometric, and psychological text features. The extracted features support the detection of fake, legitimate, and satirical news. We compared four machine learning algorithms (k-Nearest Neighbors (k-NN), Support Vector Machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGB)) to induce the detection model. Results show our proposed language-independent features are successful in describing fake, satirical, and legitimate news across three different languages, with an average detection accuracy of 85.3% with RF.

No abstract

No abstract

Currently, the widespread of fake news has raised on the political class and society members in general, increasing concerns about the potential of misinformation that can be propagated, appearing on the center of the debate about election results around the world. On the other hand, satirical news has an entertaining purpose and are mistakenly put on the same boat of objective fake news. In this work, we address the differences between objectivity and legitimacy of news documents, treating each article as having two conceptual classes: objective/satirical and legitimate/fake. Thus, we propose a Decision Support System (DSS) based on a text mining pipeline and a set of novel textual features that uses multi-label methods for classifying news articles on those two domains. For validating the approach, a set of multi-label methods was evaluated with a combination of different base classifiers and then compared to a multi-class approach. Results reported our DSS as proper (0.80 F1-score) in addressing the scenario of misleading news from challenging perspective of multi-label modeling, outperforming the multi-class methods (0.71 F1-score) over a real-life news dataset collected from several portals of news.

No abstract