MIDDAG: Where Does Our News Go? Investigating Information Diffusion via Community-Level Information Pathways

Mingyu Derek Ma,Alexander K. Taylor,Nuan Wen,Yanchen Liu,Po-Nien Kung,Wenna Qin,Shicheng Wen,Azure Zhou,Diyi Yang,Xuezhe Ma,Nanyun Peng,Wei Wang

2024-02-21

Abstract:We present MIDDAG, an intuitive, interactive system that visualizes the information propagation paths on social media triggered by COVID-19-related news articles accompanied by comprehensive insights, including user/community susceptibility level, as well as events and popular opinions raised by the crowd while propagating the information. Besides discovering information flow patterns among users, we construct communities among users and develop the propagation forecasting capability, enabling tracing and understanding of how information is disseminated at a higher level.

Social and Information Networks,Artificial Intelligence,Human-Computer Interaction

What problem does this paper attempt to address?

The paper attempts to address the problem of how to intuitively visualize and predict the dissemination paths of COVID-19 related news information on social media, and provide comprehensive insights, including the susceptibility levels of users/communities, events, and popular opinions. Specifically, the researchers developed a system called MIDDAG, which is capable of:

1. **Visualizing information dissemination paths**: Displaying the information dissemination paths at the user and community levels, allowing users to clearly see how information spreads among different users and communities.
2. **Building communities**: Grouping users into different communities based on their interaction patterns with specific news agencies, thereby analyzing the information dissemination paths at the community level.
3. **Predicting information dissemination**: Using graph neural network models to predict future information dissemination paths, helping to understand the dissemination patterns at a higher level.
4. **Assessing susceptibility**: Evaluating the susceptibility of users and communities to information (especially false information) through machine learning models, providing further insights into each dissemination path.
5. **Extracting events and opinions**: Extracting events and popular opinions from user discussions, showing whether and how core viewpoints spread.

Through these functions, the MIDDAG system aims to help researchers and the public better understand and track the process of information dissemination on social media.