Relation extraction: A deeper dive into methods for extracting information from text

An introduction to the various approaches for automatically extracting relations from text, including co-mentioning, rule-based systems, machine learning, and deep learning. You may want to watch my short introduction to the core concepts of biomedical text mining first: https://youtu.be/NcntH0WYp1M 0:00 Introduction: definition and approaches 0:42 Co-mentioning: counting, frequency normalization, strengths, and weaknesses 2:16 Rule-based: grammatical sentence parsing, hand-crafted rules, and interpretability 4:04 Machine learning: corpus annotation, text vectorization, training, and challenges 5:22 Deep learning: advantages over machine learning, pre-trained language models, BERT, and BioBERT 7:16 Use cases: systematic literature mining, databases, knowledge graphs, STRING, and other examples Hands-on exercises on text mining are available at https://jensenlab.org/training/textmining/