Ing. Jan Drchal, Ph.D.

This thesis aims to explore and develop novel methods of automated fact-checking based on machine learning NLP methods. The problem is defined as checking the veracity of a textual claim with respect to a ground truth document database. It deals with issues of large-scale document retrieval, semantic search, recognizing textual entailment, and natural language inference. The student should focus on low-resource languages such as the Czech language. An essential part of this work lies in developing methodologies for efficient dataset collection and their evaluation.