Combating fake news on social media: from early detection to intervention

Project title: Combating fake news on social media: from early detection to intervention

Project dates:  2020-2022

Grants and funding: ​ARC Discovery Project

Description

The project aims to detect fake news early to minimise the negative impact of false information. This project expects to devise novel solutions to address technical challenges for detection of fake news with scarce signals. Expected outcomes of this project include a suite of data mining and machine learning models for identification of fake news from the social media stream, prediction of user propagation of false information as well as recommendation of truthful news to counteract adversarial fake news. This project should generate technologies that enhance the integrity of the online echo system and benefit media providers and online population within Australia and across the world.

Research strategy ​  

AI and machine learning for social good.

Rationale

Australia has become a digital society, and the public always seek for news from social media platforms. Statistics show that in 2018 about 88% of the Australian population were active internet users and 79% were social media users. The wide spread of fake news becomes one of the biggest threats to the order of the Australian and international societies. We will devise computational approaches to early detection of fake news as well as recommendation of truthful news to counteract fake news. Our work directly falls within the strategic area of “Cybersecurity” with the goal of enhancing the integrity and credibility of online information. The technology

produced in this project has wide social and cultural benefits to the Australian community, and will directly benefit Australian government agencies and companies that provide social media surveillance services for law enforcement and other applications.

Key people

• A/Prof Xiuzhen Zhang, Prof Yan Wang (Macquarie University)
• Prof Huan Liu (Arizona State University, USA)

Associated journal publications

• Tian, L., Zhang, X., Wang, Y. and Liu, H., 2020, April. Early detection of rumours on Twitter via stance transfer learning. In European Conference on Information Retrieval (pp. 575-588). Springer, Cham.
• Tian, L., Zhang, X. and Peng, M., 2020, April. FakeFinder: Twitter Fake News Detection on Mobile. In Companion Proceedings of the Web Conference 2020 (pp. 79-80).