News ScanNER: Entity Tagging in News Headlines | Deep Learning Workshop Capstone
Because news organizations produce hundreds of articles and stories every day, they run into many problems that can be addressed using named entity recognition (NER). NER tags, such as the names of people, organizations, and locations, can be used to tag text data from articles to create metadata for grouping articles, building recommendations, improving searches, and keeping up with trends. Our NER web app uses a BERT-based architecture to tag people, organizations, locations, and miscellaneous other entities. The web app also pulls and tags recent headlines from CNN’s API, highlighting the entities in each of CNN's main content categories.