Named entity recognition: A deeper dive into methods for finding things mentioned in papers

An introduction to dictionary-based and machine learning-based named entity recognition. The presentation gives an overview of how to create a dictionary, how to use one for recognizing entities in text, and how to complement this with machine 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 subtasks 1:09 Dictionary: unique identifiers, biomedical databases/ontologies, and name expansion 3:01 Dictionary-based NER: flexible matching, performance, and dealing bad names 4:58 Machine learning: name shape, textual context, annotated corpus, and deep learning 7:19 Use cases: curation support, concept indexing, and relation extraction An overview of text-mining resources is available at https://jensenlab.org/resources/textmining/ and hands-on exercises at https://jensenlab.org/training/textmining/