IARA: an architectural model to support the development of proactive chatbots for automated misinformation detection and media literacy education

Abstract

In the last few years, the widespread dissemination of misinformation through social platforms has become a critical issue. In several developing countries, such as Brazil, India, and Mexico, instant messaging apps, including WhatsApp and Telegram, have been one of the main sources of misinformation. These apps offer a significant resource: public groups. However, many of these groups are used to spread misinformation, especially as part of well-organized political or ideological campaigns. In this paper, we propose an architectural model, called IAra, to assist in the development of pro-active chatbots for the automatic detection of misinformation and media education. Additionally, following the proposed architectural model, we have developed a proactive chatbot called IAraBot, which detects the presence of misinformation automatically and in real-time. Moreover, the IAraBot provides hints and practical examples for users to develop the necessary skills to critically identify content containing misinformation. The IAraBot was developed for WhatsApp and Telegram platforms.