Two publications accepted to Hypertext and SIGIR workshop ROME

2019-10-07T21:41:32+01:00July 9th, 2019|Categories: Outcomes, Publications|

Fake News Reading on Social Media: an Eye-tracking Study

Abstract: The online spreading of fake news (and misinformation in general) has been recently identified as a major issue threatening entire societies. Much of this spreading was enabled by new media formats, namely social networks and online media sites. Researchers and practitioners have been trying to answer this by characterizing the fake news and devising automated methods for detecting them. The detection methods had so far only limited success, mostly due to the complexity of the news content and context and lack of properly annotated datasets. One possible way to boost the efficiency of automated misinformation detection methods, is to imitate the detection work of humans. In a broader sense of dealing with fake news spreading, it is also important to understand the news consumption behavior of online users. In this paper, we present an eye-tracking study, in which we let 44 participants to casually read through a social media feed containing posts with news articles. Some of the presented articles were fake. In a second run, we asked the participants to decide on the truthfulness of these articles. We present the description of the study, characteristics of the resulting dataset (which we hereby publish) and several findings.

Monant: Universal and Extensible Platform for Monitoring, Detection and Mitigation of Antisocial Behaviour

Abstract: Growing negative consequences of online antisocial behaviour have recently elicited many research efforts, aimed at mitigating or even eliminating this undesired behaviour. However, addressing the open problems is challenging (among other) due to lack of suitable datasets. Also, platforms, where the research results may be applied, are missing too. Therefore, we propose a universal and extensible platform named Monant. It is specifically designed to support characterization and detection of multiple types of antisocial behaviour. Monant does so by means of collecting multimodal, multilingual context-rich data from multiple types of web sources. In addition, the platform supports the deployment of various novel mitigation tools, where data-driven approaches can be applied. To demonstrate the unique characteristics of our platform, we conducted an experimental task in which we monitored healthcare misinformation and identified the most frequent false medical claims related to cancer treatment. Finally, we describe several use cases, which are feasible in our platform and which correspond to trending research directions.