Causal Knowledge Graphs for Counterfactual Claims Reasoning
AI approaches to misinformation detection.
Current AI approaches to misinformation detection often learn to recognise paraphrases of previously seen claims. Detecting new misinformation is much harder, and linguistic cues are not enough to distinguish fact from fiction. Our approach is grounded in knowledge graphs and the logic of causality. However, this approach has its own challenges. Much of the misinformation encountered is not limited to simple factual statements that can be tested against a structured representation of knowledge but it consists of more complex claims such as counterfactual statements (e.g. “this would never have happened if…”). To address this problem, we integrate approaches from different subfields of computer science, namely, computational logic, deep learning and natural language processing.
- Ameer, Saadat-Yazdi; Xue, Li; Sandrine, Chausson; Vaishak, Belle; Björn, Ross; Jeff Z., Pan; Nadin, Kökciyan (2022). KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. The 9th Workshop on Argument Mining Special Theme: Argument Mining in Real-World Applications 2022.
- Chausson, S., Saadat-Yazdi, A., Li, X., Pan, J. Z., Belle, V., Kokciyan, N., & Ross, B. (Accepted/In press). A Web-based Tool for Detecting Argument Validity and Novelty. In The International Conference on Autonomous Agents and Multiagent Systems (AAMAS) Demonstrations Track 2023 ACM.
- Jung, Anna-Katharina; Ross, Björn; Stieglitz, Stefan (2020). Caution: Rumors ahead—A case study on the debunking of false information on TwitterBig Data & Society 7(2). [doi]
- Ross, Björn; Jung, Anna-Katharina; Heisel, Jennifer; Stieglitz, Stefan (2018). Fake News on Social Media: The (In)Effectiveness of Warning Messages. In Proceedings of the International Conference on Information Systems (ICIS), San Francisco, December 2018.
- V. Belle (2022). Tractable Probabilistic Models for Ethical AI. ICCS.
- Xue, Li; Alan, Bundy (2022). An overview of the ABC Repair System for Datalog-like Theories, The 3rd International Workshop on Human-Like Computing.
- “Detecting False Claims about Climate Change: An Approach based on Knowledge Graphs”. Xue Li (Delivered at Data Science 4 Sustainable Development: Workshop)
False claims on social media are harmful which could influence individuals’ opinions and actions. As climate change is a global crisis which requires collaboration across our whole society, detecting false claims in that domain is particularly essential. In this talk, we propose a system that detects false claims in the domain of climate change by querying KGs, where probabilistic logic programming will be employed. This system aims at classifying a given claim as being supported or refuted with a probability and explanations.