Towards Explainable Fact Checking

Abstract

The past decade has seen a substantial rise in the amount of mis- and disinformation online, from targeted disinformation campaigns to influence politics, to the unintentional spreading of misinformation about public health. This development has spurred research in the area of automatic fact checking, from approaches to detect check-worthy claims and determining the stance of tweets towards claims, to methods to determine the veracity of claims given evidence documents. These automatic methods are often content-based, using natural language processing methods, which in turn utilise deep neural networks to learn higher-order features from text in order to make predictions. As deep neural networks are black-box models, their inner workings cannot be easily explained. At the same time, it is desirable to explain how they arrive at certain decisions, especially if they are to be used for decision making. While this has been known for some time, the issues this raises have been exacerbated by models increasing in size, and by EU legislation requiring models to be used for decision making to provide explanations, and, very recently, by legislation requiring online platforms operating in the EU to provide transparent reporting on their services. Despite this, current solutions for explainability are still lacking in the area of fact checking. This thesis presents my research on automatic fact checking, including claim check-worthiness detection, stance detection and veracity prediction. Its contributions go beyond fact checking, with the thesis proposing more general machine learning solutions for natural language processing in the area of learning with limited labelled data. Finally, the thesis presents some first solutions for explainable fact checking.

arXiv commentThesis presented to the University of Copenhagen Faculty of Science in partial fulfillment of the requirements for the degree of Doctor Scientiarum (Dr. Scient.)

References in corpus (51)

• Scikit-learn: Machine Learning in Python
• Efficient Estimation of Word Representations in Vector Space
• Distributed Representations of Words and Phrases and their Compositionality
• Distilling the Knowledge in a Neural Network
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Decoupled Weight Decay Regularization
• Natural Language Processing (almost) from Scratch
• Axiomatic Attribution for Deep Networks
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• European Union regulations on algorithmic decision-making and a "right to explanation"
• Frustratingly Easy Domain Adaptation
• Theano: new features and speed improvements
• CTRL: A Conditional Transformer Language Model for Controllable Generation
• SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents
• Model-Agnostic Interpretability of Machine Learning
• A review of domain adaptation without target labels
• Sanity Checks for Saliency Maps
• Computational fact checking from knowledge networks
• Automatic Detection of Fake News
• Ask the GRU: Multi-Task Learning for Deep Text Recommendations
• Clustered Multi-Task Learning: A Convex Formulation
• e-SNLI: Natural Language Inference with Natural Language Explanations
• Pathologies of Neural Models Make Interpretations Difficult
• Learning Task Grouping and Overlap in Multi-task Learning
• Know What You Don't Know: Unanswerable Questions for SQuAD
• BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
• A simple but tough-to-beat baseline for the Fake News Challenge stance detection task
• The (Un)reliability of saliency methods
• Learning Reporting Dynamics during Breaking News for Rumour Detection in Social Media
• Discourse-Aware Rumour Stance Classification in Social Media Using Sequential Classifiers
• Diachronic word embeddings and semantic shifts: a survey
• Positive-Unlabeled Learning with Non-Negative Risk Estimator
• XFake: Explainable Fake News Detector with Visualizations
• All-in-one: Multi-task Learning for Rumour Verification
• How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation
• Bayesian Multitask Learning with Latent Hierarchies
• Detecting Cross-Modal Inconsistency to Defend Against Neural Fake News
• DeClarE: Debunking Fake News and False Claims using Evidence-Aware Deep Learning
• Popularity of arXiv.org within Computer Science
• Deep Multitask Learning for Semantic Dependency Parsing
• Adversarial Domain Adaptation for Stance Detection
• Recurrent Neural Network-Based Sentence Encoder with Gated Attention for Natural Language Inference
• On the Importance of Delexicalization for Fact Verification
• Time-Aware Evidence Ranking for Fact-Checking
• Distantly Supervised Named Entity Recognition using Positive-Unlabeled Learning
• Universal Adversarial Perturbation for Text Classification
• Semantics Preserving Adversarial Learning
• Combining Sentiment Lexica with a Multi-View Variational Autoencoder
• 2kenize: Tying Subword Sequences for Chinese Script Conversion
• Neural Check-Worthiness Ranking with Weak Supervision: Finding Sentences for Fact-Checking
• Other Topics You May Also Agree or Disagree: Modeling Inter-Topic Preferences using Tweets and Matrix Factorization