Named entity recognition (NER) is a key component of many scientific literature mining tasks, such as information retrieval, information extraction, and question answering; however, many modern approaches require large amounts of labeled training data in order to be effective. This severely limits the effectiveness of NER models in applications where expert annotations are difficult and expensive to obtain. In this work, we explore the effectiveness of transfer learning and semi-supervised self-training to improve the performance of NER models in biomedical settings with very limited labeled data (250-2000 labeled samples). We first pre-train a BiLSTM-CRF and a BERT model on a very large general biomedical NER corpus such as MedMentions or Semantic Medline, and then we fine-tune the model on a more specific target NER task that has very limited training data; finally, we apply semi-supervised self-training using unlabeled data to further boost model performance. We show that in NER tasks that focus on common biomedical entity types such as those in the Unified Medical Language System (UMLS), combining transfer learning with self-training enables a NER model such as a BiLSTM-CRF or BERT to obtain similar performance with the same model trained on 3x-8x the amount of labeled data. We further show that our approach can also boost performance in a low-resource application where entities types are more rare and not specifically covered in UMLS.

For the North American hurricane season, in partnership with the Tennessee Valley Authority (TVA) and working with the U.S. Department of Energy, Office of Electricity Delivery and Reliability, we have developed a capability that helps visualize the status of the electric transmission system infrastructure. The capability toolkit, called VERDE - Visualizing Energy Resources Dynamically on Earth, takes advantage of the Google Earth^reg platform to display spatio-temporally informed power grid and related data. Custom libraries describe the electrical transmission network in the Eastern United States and the dynamic status of each transmission line. Standard Google Earth layers provide rich spatial context. In addition to live status, VERDE provides a framework and mechanism to ingest predictive models, data from different sources, and response.